lINk TO CIGARETTE SMOkE

Lung cancer remains the most common fatal malignancy in the Western world. Survival rates have only improved modestly over the past three decades and new approaches are urgently required. It is clear that a concerted effort to reduce cigarette smoking is required. However, about 10% of patients with lung cancer are never smokers, indicating genetic or other predisposition. Lung cancer screening programmes are being trialled to target high-risk populations. Genetic strategies will provide new methods for screening and predicting response to treatment. Current therapy for lung cancer has reached a plateau and novel agents have shown modest clinical efficacy. Understanding the mechanisms by which chronic inflammatory disorders such as chronic obstructive pulmonary disease contribute to lung cancer development will help to identify new biological targets and biomarkers of early disease. This review focuses on recent advances in lung cancer prevention and treatment. kEywORDS Biological therapy, genetics, inflammation, lung cancer, screening, smoking DEClARATION Of INTERESTS No conflict of interests declared. Symposium review 1PS Hodkinson, 2T Sethi 1Clinical Lecturer in Respiratory Medicine, University of Edinburgh; 2Professor of Respiratory Medicine, King’s College Hospital, London, UK Correspondence to T Sethi, Department of Respiratory Medicine and Allergy, King’s College Hospital, Denmark Hill Campus, Bessemer Road, London SE5 9RS, UK tel. +44 (0)20 3299 3165 e-mail [email protected] to lung cancer prevention and to reducing the risk of recurrence after curative treatment. Advances in smoking cessation Cigarette smoke-related cancer remains a significant health burden, despite changes in government policy and public health campaigns. This is in part due to nicotine addiction promoting continued smoking and thus exposure to inhaled carcinogens. Until recently, smoking cessation therapy has focused on counselling, nicotine replacement and bupropion. Unfortunately, long-term cessation rates are poor (approximately twice placebo). Recent work has advanced our knowledge of nicotine addiction and produced targeted drug therapy to promote smoking cessation. In 2006, a clinical trial compared varenicline, a partial agonist at the a2-b4 nicotinic receptor, with sustained release bupropion or placebo. Importantly, 12 weeks of varenicline therapy produced significantly greater continuous abstinence rates compared with the other two interventions. An additional study demonstrated that prolonged therapy with varenicline (24 weeks) improved abstinence rates compared with placebo, even when the drug was stopped, although long-term quit rates remained low. These studies resulted in the approval of varenicline for use in smoking cessation. It is also important to note the key role of smoking cessation services in coordinating cessation therapy and improving quit rates. It is clear that further research is required into the mechanisms of nicotine addiction. A new report highlights a nicotine addiction locus on chromosome 15q24-25, which includes the a5-a3-b4 nicotinic receptor gene cluster. A polymorphism, which alters an amino acid in this nicotinic receptor, is associated with lung cancer and chronic obstructive pulmonary disease (COPD). It is hoped that this kind of research will identify novel genes that promote nicotine addiction and thus reveal new targets for drug therapy. The future may involve assessing ‘addiction’ genes in smokers to allow individualised therapy. GENETIC RISk fACTORS Approximately 10% of lung cancers arise in life-long nonsmokers. This group accounts for nearly 3,000 deaths each year in the UK. The development of lung cancer in non-smokers cannot be solely attributed to passive smoking or atmospheric pollution, suggesting the existence of genetic and/or other risk factors. This hypothesis is supported by the observation that smokers who have a first-degree relative diagnosed with lung cancer before 50 years of age have a higher risk of developing lung cancer than those with no family history. Bailey-Wilson and colleagues’ linkage analysis of 52 families with three or more individuals affected by respiratory tract malignancy showed that susceptibility to lung cancer maps to a locus on chromosome 6q2325. Detailed mapping of this region revealed a potential candidate gene, although its function is unknown. Genome-wide association studies have also identified loci on chromosomes 5p15, 6p21 and 15q25 that are associated with lung cancer. Of particular interest, 15q25 contains three cholinergic nicotine receptor genes and variations in these genes may contribute to lung cancer risk independent of effect on nicotine addiction. These loci only account for 10% of familial lung cancer risk and stratification of current and ex-smokers by number of risk variants shows that smoking is the predominant risk factor. Other approaches to identify genetic factors associated with lung cancer have focused on differential gene expression profiles from airway epithelium of current and never smokers and the sequencing of lung cancer genomes. It is anticipated that research of this type will identify at-risk populations for screening, define markers of early disease and suggest novel targets to prevent the carcinogenic effects of tobacco smoke. lUNG CANCER SCREENING Screening for lung cancer is an attractive option. Symptoms caused by lung cancer occur at a late stage as the tumour invades local structures or spreads to distant sites. Thus, the majority of patients present with advanced disease. Early-stage lung cancer has a better survival rate, primarily as a result of response to treatment and therefore identification of small lung tumours before symptoms are present may greatly affect lung cancer survival. Several challenges need to be addressed before a successful lung cancer screening programme can be established: defining a screening method, determining the population for screening and developing a pathway for follow-up of pulmonary nodules. Chest X-ray and computed tomography screening Initial screening programmes in lung cancer using chest radiographs and/or cytological analysis of sputum in male smokers have produced disappointing results. Computed tomography (CT) screening may be a better technique as it is able to detect lung parenchymal changes before they are apparent clinically or on chest X-ray. Several lung cancer screening trials using low-dose CT showed that malignant tumours can be detected at an earlier stage than by clinical assessment. However, these trials were not controlled in design and therefore effect on mortality is difficult to determine. In addition, these studies showed high false positive rates and low rates of detection of incident cases of lung cancer. To address these issues, randomised controlled trials of low-dose CT in lung cancer screening are being conducted. Designed to compare the effect of screening with low-dose CT or chest X-ray on lung cancer Advances in lung cancer J R Coll Physicians Edinb 2011; 41:142–9 © 2011 RCPE 143 edcation mortality, the largest of these studies recruited more than 50,000 patients in the US between 2002 and 2004. The last round of screening was carried out in 2007 and an overview report has been published this year. Smaller European studies have reported preliminary results on prevalence of lung cancer at baseline screening but have yet to complete. Screening programmes are not without inherent difficulties, some of which are specific to lung cancer. Many of the reported screening studies show that uptake by the ‘at-risk’ population is low and methods for recruitment are inefficient. In addition, the prevalence of pulmonary nodules detected by CT is high in the ‘at-risk’ population. However, the majority of nodules are benign, even in smokers, raising concerns about over-investigation of potential malignancy. Finally, although some screening studies have suggested improvements in lung cancer mortality in the screened population, other investigators have suggested that this may simply be due to detection of cancer earlier within its natural history or so-called ‘lead-time’ bias. UK lung screening trial To address these issues Baldwin et al. have recently described the UK lung screening trial (UKLS). This will recruit 28,000 patients from seven centres in the UK with ≥5% lung cancer risk within five years, determined using a well-established model of lung cancer risk (Liverpool Lung Project risk model). Patients will be randomised to either low-dose CT screening or no screening with a ten-year follow-up. The study will employ a single-screen design with a nodule management protocol based on volumetric analysis and nodule characteristics, which determines referral to the multidisciplinary team or CT follow-up. It also defines a new protocol for the follow-up of pulmonary nodules. It is hoped that this study will not have the problems of longterm compliance as it employs a single screen technique and may prove to be cost-effective as it targets a highrisk population. It is highly likely that routine CT screening in the UK will wait until the results of this trial. ADVANCES IN THE STAGING Of NSClC The primary influence on lung cancer treatment decisions by the multidisciplinary team is the stage of the tumour at diagnosis. The Tumour Node Metastasis (TNM) classification of NSCLC, which originated in 1974, has recently been revised. For the first time, this revision (the seventh edition) is based on analysis and validation of outcome data from a large international database of lung cancer cases. There have been several important alterations in the descriptions of the T and M groupings to reflect differences in patient prognosis, which are summarised in Table 1. Following careful analysis, the N descriptor has remained unchanged (N0–3). Although these new TNM subgroups have resulted in a more complex definition of the stage groups (I–IV), the seventh edition classification is widely accepted as a more evidencebased system for the basis of treatment decisions. Staging techniques for lung cancer have also advanced in the past decade, including more sophisticated imaging such as 18F-deoxyglucose positron emission tomography (FDGPET). This technique is now widely available to cancer centres and has been shown to accurately stage lung cancer and prevent unnecessary radical treatment. However, its applicability is limited by false positive and negative results and confirmation by histology is often necessary. Minimally invasive staging techniques have progressed significantly, particularly in the field of endobronchial ultrasound (EBUS). Endobronchial ultrasound allows almost complete staging of mediastinal lymph nodes, can be carried out as a day case and has a high sensitivity and specificity. It is now often the first-choice diagnostic and staging technique for suspected NSCLC with an involvement of the mediastinal nodes on imaging (CT or FDG-PET) and can prevent unnecessary thoracotomy. Endobronchial ultrasound is, however, limited to fine needle aspirate (FNA) samples, which may not be sufficient for the assessment of biomarkers of treatment response. This issue is being addressed by research on ribonucleic acid extracted from EBUS-FNA samples. ADVANCES IN CHEMOAND RADIOTHERAPy