At the onset of the 21st century, cancer of the lung remains the most common cause of cancer deaths in men and women alike, as it has been for many years (Landis et al, 1999). The link between lung cancer and cigarette smoking was emphasized half a century ago by Wynder and Graham (1950) and became officially recognized in 1957 when the Surgeon General of the United States, Leroy E. Burney, issued a statement declaring, “Excessive smoking is one of the causative factors in lung cancer.” This was followed by a Public Health Service Monograph, “Smoking and Health” published in 1964 (PHS monograph 1103) that exhaustively reviewed the health effects of cigarette smoking and clearly established the relationship between cigarette smoking and lung cancer. It may be assumed that a subsequent decrease in cigarette smoking accounts for the recent modest drop in the incidence and deaths from this disease in the US. The relationship between cigarette smoking and lung cancer is complex, however, and individuals differ in their susceptibility to the carcinogenic effects of cigarette smoking and probably other environmental agents (Spitz, 1999); the challenge of the future will be to identify those who are constitutionally at risk.

Lung cancer is also an occupational disease. The earliest recorded cases of occupational lung cancer, first recognized in 1879, were among the Schneeberg and Joachimstal miners of Czechoslovakia who were exposed to radon gas in material known as Pitchblende, from which Pierre and Marie Curie extracted and isolated radium. In the US, lung cancer in Colorado uranium miners has been attributed to radiation (Archer et al, 1974; Saccomanno et al, 1988). Arsenic, long known to produce cutaneous hyperkeratoses, is now also recognized as a cause of lung cancer. At least 12 substances found in the workplace are considered to be lung carcinogens in humans, and 5% of lung cancers in the US have been attributed to occupational exposure (Doll and Peto, 1981). The industrial agents reported to cause lung cancer include chloromethyl ether, mustard gas, polycyclic aromatic hydrocarbons, crystalline silica, nickel, chromium, beryllium, cadmium, and asbestos, the last in association with cigarette smoking (Braun and Truant, 1958; Cordova et al, 1962; Talcott et al, 1989; Rosenman and Stanbury, 1996; Steenland et al, 1996; Beckett, 2000). Of these, asbestos is of particular interest to the cytopathologist because these fibers can be identified in specimens of sputum (see Chap. 19). Asbestos fibers may be observed in sputum simultaneously with cancer cells, as was first demonstrated by An and Koprowska in 1962. For a comprehensive review of the pathology of asbestos associated diseases, see Roggli et al, 1992.

Finally, human papillomavirus (HPV) has been implicated in the pathogenesis of squamous lung cancer. Syrjänen et al (1989) reported finding HPV types 6 and 16 in 9 of 131 squamous lung cancers. On the other hand, Stremlau et al (1985) hybridized tissue from 24 lung tumor biopsies with 10 different HPV types and found only 1 carcinoma with HPV (type 16). The latter occurred in a woman with a history of treated cervical carcinoma and the lung tumor may have represented metastatic cancer. Thus, the evidence that HPV has a role in carcinogenesis of the lung is unconfirmed. See also comments on the role of HPV in cervical cancer (Chap. 11), lesions of the oral cavity and larynx (Chap. 21), and the esophagus (Chap. 24).

The relative frequency of the histologic subtypes of lung cancer in the US has changed dramatically over the last several decades (Vincent et al, 1977; Johnston, 1988; Devesa et al, 1991; Sobue et al, 1999). Squamous cancers, which were predominant in the 1950s and 1960s, now account for no more than 30% of cases. At the same time, peripheral adenocarcinomas of the lung have increased in frequency and are now the most common type of lung cancer in the US. The reason for this is unknown but has been attributed to changes in the manufacture of cigarettes, the use of filter tips, or the possible effects of other environmental cofactors. As a result, sputum cytology, which best detects early squamous carcinomas, is less useful as a screening tool now than in prior years. On the other hand, percutaneous aspiration of peripheral lung lesions has become increasingly important in early diagnosis of small peripheral carcinomas, even as tiny as 2 to 5 mm, and may prove of value in assessing lesions found by high-resolution spiral or helical computed tomography (CT) (Henschke et al, 1999).

Surgery remains the treatment of choice for all but small-cell lung cancer; even with recent advances in chemotherapy and radiotherapy, the best opportunity for long-term survival and cure of lung cancer lies in early diagnosis and surgical resection. A major feasibility study of early lung cancer detection, encompassing over 30,000 cigarette-smoking men who were followed for 5 to 8 years (Berlin et al, 1984; Flehinger et al, 1984; Fontana et al, 1984; Frost et al, 1984) has now been concluded (Melamed et al, 1984; Melamed and Flehinger, 1987; Flehinger et al, 1988; Melamed, 2000) and is summarized below. The results of this study failed to show a decrease in death rates from lung cancer, leading the American Cancer Society and other influential agencies to recommend against screening for lung cancer. Chest x-rays and sputum cytology were considered to be diagnostic tools for symptomatic patients only. Because symptomatic lung cancer is usually advanced lung cancer, however, the opportunity for early diagnosis and cure is lost. In a recent evaluation of the statistical basis for this recommendation, Dempster (1998) concluded that the study data “strongly support a finding
of benefit from routine screening of high-risk populations.” Flehinger and Kimmel (1987) have estimated that annual radiographic screening alone would decrease mortality from adenocarcinoma of the lung by up to 18%. The results of extensive Japanese lung cancer studies also suggest that surgical removal of very small adenocarcinomas, discovered by the new methods of CT, can improve the cure rate significantly (Noguchi et al, 1995).

Ongoing studies of lung cancer detection, using spiral CT (spiral or helical CT), known as the Early Lung Cancer Action Project (ELCAP), summarized later in this chapter, may yet modify future recommendations regarding screening for lung cancer (Henschke, 2000). This project, if successful, is likely to pose new challenges for interpretation of cytology of sputum, bronchial brushes, bronchoalveolar lavage, and particularly transbronchial and percutaneous needle aspirates.

It must be emphasized that cytology is a method of choice in the diagnosis of radiologically detected lung lesions suspected of being malignant. Thus, it is appropriate to analyze briefly the existing methods of diagnosis of lung cancer with attention to the role of cytology.