1941- 50 I 1930 194G 1950 96G 1970 1980 YEAR 70- 60- 50% 5 40. Ii! &O- 20 10 ?m - l! FIGURE la--Changes in the prevalence of cigarette smoking among successive birth cohorts of men, 1900-1978 NOTE Calculated from the results of over 13,ooO intervwws conducted during the last two quarters of 1978. Provided by the Division of Health Interview Statistxs, U S National Center for Health Statistics. SOURCE: U.S. Department or Health, Education, and Welfare wfa A third concern about the coherence of smoking behavior and lung cancer mortality has been that overall lung cancer mortality continues to rise at a time when the prevalence of cigarette smoking continues to decline, and the consumption of lower tar and nicotine cigarettes is increasing. Part of this apparent discrepancy can be accounted for by the relatively slow decline in the excess risk of 51 I I I I 1 1 30. 40- 50. so- 70- 80. AGE FIGURE 13.-Age-specific mortality rates for cancer of the bronchus and lung, by birth cohort and age at death for males, United States, 1950-1975 SOURCE: Derived from data available in National Cancer Institute (19% 52 10 n YEAR FIGURE 14.-Changes in the prevalence of cigarette smoking among successive birth cohorts of women, 1900-1978 ~O'E3: Calculated from the results of over 13,ooo interviews conducted during the last two quarters of 1978. Pmvided by the Division of Health Interview Statistics. U.S. National Center for Health Statistics. SOURCE: U.S. Department of Health, Education, and Welfare (200). developing lung cancer once someone actually stops smoking, compared to persons who continue to smoke cigarettes. However, in the youngest male birth cohorts (birth years 1931-1940 and 1941- 1950), there is a substantially lower peak prevalence of smoking which should result in a lower lung cancer mortality experience. From the smoking prevalence data and Figure 12, one would expect to see this declining mortality experience in those birth cohorts born after 1930, and the data in Figure 13 for 1935 and 1940 birth cohorts suggest that a decline in mortality experience is occurring. This trend can be visualized easily in Figure 16, which plots the age- specific lung cancer mortality rates for 5-year age groups over time, and reveals that the male rates for the youngest age groups do appear to be declining. No such trend can be seen in the female mortality experience, and this, too, is consistent with the smoking Prevalence data presented in Figure 14. 53 54.0 10 0.5 04 03 0.2 I I I I 1 r- 30. 40. 50. 60. 70. Eo- AGE FIGURE 15.-Age-specific mortality rates for c+ncer of the bronchus and lung, by birth cohort and age at death for females, United States, 1956-1975 SCIJRCE: Derived fmm data avaiiable in National Cancer Institute (198). 54 TABLE 13.-Lung cancer mortality ratios for male and female smokers at 6- and la-year followup, ACS 25-State Study sex NOll- smokers Mortality ratios By&W folkJwup 12.year followup Males 1.00 9.20 8.53 Females 1.00 2.20 3.58 When the prevalence of cigarette smoking by birth cohort is compared with the mortality experience by birth cohort, the relationship between cigarette smoking behavior and lung cancer mortality experience is extremely coherent. This is also supported when lung cancer mortality ratios are examined at various periods of foliowup in the prospective studies. In the ACS 25-State Study, a different pattern of lung cancer mortality emerges for males compared to females. In contrast to lung cancer mortality ratios among male smokers, which remained almost constant during the g-year followup interval, ratios for female smokers increased (Table 13). A similar trend is observed among male U.S. Veterans as noted above for males in the ACS 25-State Study. Figure 17 presents lung cancer mortality ratios by amount smoked for male veterans at 8'1, years compared to 16 years' followup. No differences between the two periods are evident and the pattern is constant at each level of exposure. Lung Cancer Mortality and Premalignant Changes in Bronchial. Epithelium Since smoking is significantly associated with lung cancer, smok- ers could be expected to develop premalignant changes in bronchial epithelium more commonly than nonsmokers prior to the develop- ment of frank cancer. In the late 195Os, one scientist (9, 14, 15) examined the tracheobronchial tree of 402 malesat post mortem in a controlled blinded study and ,found that several kinds of changes were much more common in the tracheobronchial tree of smokers as compared with nonsmokers (Table 14). The frequency and intensity of these epithelial changes (loss of cilia, basal cell hyperplasia, presence of atypia) correlated with the number of cigarettes smoked. The most severe lesions, aside from invasive cancer, were not seen among males who did not smoke regularly and were found only rarely among light smokers. They were present, however, in 4.3 .percent of sections from males who smoked one to two packs a day, 55 l.ooOO 9030 I9330 700.0 6000 E YSJO 4000 F XXI0 2000 loo0 900 600 70.0 1 >- 600 - 500 - g 400 - 4 ii 30.0 - i 200 - mz 2 c" 2 10.0 - 90 - 80 - 7.0 - 6.0 - 50 - 40 - 3.0 - 20 - 10 - 09 - 06 - 07 - 06 - 05 - WHlTE MEN 5054 *err FIGUflE 16.-Mortality rates for malignant neoplasm of the trachea, bronchus, and lung, for white men and white women, by birth cohort and age at death, United States, 5-year intervals during 1947-1977 SOURCE!National Center for Health Statistics 1200). 56 WWTE WOMEN ICOO - So0 - WI0 - 700 - 600 - 500 - 400 - 3Jo - M.0 - 100 - so - 6.0 - 70 - 60 - 50 - 40 - 30 - 20 - FIGURE 16, continued.-Mortality rates for malignant neoplasm of the trachea, bronchus, and lung, for white men and white women, by birth cohort and age at death, United States, Byear intervals during 1947-1977 SOURCE. Natmnal Center for Health StatistmG?GOI. 57 16 years NON- SMOKERS I I I I I I I I I I I I 0 2 4 6 0 10 12 14 16 18 20 22 24 MORTALITY RATIO FIGURE 17.-Lung cancer mortality ratios for male smokers by amount smoked, S'/,- and l&year followup, U.S. Veterans Study in 11.4 percent of sections from males who smoked two or more packs a day, and in 14.3 percent of sections from smokers who died of lung cancer. Studies by the same authors and others (7, 10, 28, 39, 51, 89, 96, 144, 206, 217, 233, 268,298, 319) have confirmed this relationship between smoking and premalignant changes in bronchial epithelium in males and females, with and without lung cancer. More recent investigations (121, which examined the histologic changes in the bronchial epithelium of male cigarette smokers who had died from causes other than lung cancer, found that changes occurred far less frequently in nonsmokers than in cigarette smokers. Changes in smokers correlated with the amount smoked. When comparing the degree of histologic changes of men who died in 58 TABLE 14.-Percent of slides with selected lesions,8 by smoking status and presence of lune cancer Group Number Number CaSeS slides Percent of slides with cilia absent and averaging 4 or more cell rows in depth No cel!s Some cells All cells atypical atypical atypic& Total Cases without lung cancer Never smoked regularly 65 Excigarette smokers 72 Cigarettes-`/, pk. a day 36 Cigarettes-V-1 pk. a day 59 Cigarette-l-2 pks. a day 143 Cigarettes-2+ pks. a day 36 Lung cancer cw 63 3,324 1.0 0.03 - 1.1 3,436 3.5 0.4 0.2 4.1 1.624 0.2 4.2 0.3 4.1 3,016 - 7.1 0.8 7.9 7.062 12.6 4.3 16.9 1,787 - 26.2 11.4 37.5 2,784 - 12.5 14.3 26.8 "In some sectmns. two or more lesmns were found. In such mstances. all of the lesions were counted and are Included in both mdiwdual columns and in the total column of the table. Lesions found at the edge of an uicer were excluded "These laions may be call& carcmoma m-situ. `Of the 63 who died of lung cancer. 55 regularly smoked cigarettes up to the time of diagnosis. 5 regularly smoked cigarettes but stopped before diagnosis. 1 smoked cigars. 1 smoked pipe and cigars. 1 was an occasional cigar smoker. SOURCE. Auerbach 19. 14. 151 the period 1955-1960 with those who died in 1970-1977, these investigators found the latter exhibited less advanced histologic changes. The authors attributed this finding to the reduced tar and nicotine yield of cigarettes smoked by this group when compared to the average tar and nicotine yield of those smoked by the earlier group (Table 15). Several investigators have examined the relationship between smoking and cytological changes in respiratory epithelial cells shed into sputum in groups of smokers and nonsmokers. These studies (171, 193, 220, 262) have generally found increased proportions of sputum specimens showing atypical cells among smokers as com- pared with nonsmokers, and these changes have progressed toward cancer with increasing duration of the smoking habit. In addition, these changes have reverted toward normal in individuals who stopped smoking. These data support the causal nature of the association between smoking and lung cancer. Experimental Studies Over the past 30 years, a number of experimental models have been developed to study tobacco-induced carcinogenesis. These data are explored in detail in the Part of this Report on the mechanisms of carcinogenesis. Lung Cancer and Non-Cigarette Tobacco Use The relationship between lung cancer and other forms of tobacco was comprehensively reviewed in reports by the U.S. Public Health 59 377-310 0 - 82 - 6 TABLE l&-Percentage of sections with each of several categories of histologic change, classified according to smoking habit * Adjusted % Adjusted % Adjusted % Adjusted % Never Smoked Smoked 1-19 Smoked 2CX39 Smoked 40+ HiSt&giC Regularly Cigarettes/ Cigarettes/ Cigarettes/ change hY DRY Day A B A B A B A B Basalcell hyperplasia: Total 6+ rcnvs IO%+ cells with atypical nuclei 30%+ cells with atypical nuclei 50%+ cells with atypical nuclei 70%+ cells with atypical nuclei Lesion with cilia absent: T&d lo%+ cells with atypical nuclei 30%+ cells with atypical nuclei 50%+ cells with atypical nuclei 70%+ cells with atypical nuclei 100% cells with atypical nuclei No. of sections No. of subjects 3.8 5.8 0 0.1 0.1 0.5 0.1 0.4 0 0.1 0 0 5.3 4.2 0 ,8 cigars per day.. . . . . Pipe smokers: < 5 pipefuls per day.. . . . . . . . . . . 5 to 19 pipefuls par day . . . . . . . > 19 pipefuis per day . . . . . . . . . Cigar and pipe: 8 or lean cigars, 19 or leas pipefuls . . . . . > 8 cigars, > 19 pipefuls . . . . . . . . . . . . . 12 11 2 1.00 78 1.14 264 207 .77 220 247 2 12 3 1.62 18 219 2 SOURCE: Kahn (131). response relationship; however, the relationship is not as strong as that noted for cigarett.e smoking. A few retrospective studies contain adequate numbers of smokers tc allow an examination of dose-response relationships between pipe and cigar smoking and lung cancer (I, 161, 215, 230). An increased risk for developing lung cancer correlated with the increased use of pipes and cigars as measured by amount smoked and depth of inhalation. 61 Several investigators have examined histological changes in lungs of cigar and pipe smokers. One study (15) examined 36,340 histologic sections for various epithelial lesions obtained from 1,522 white adults. The numbers and types of pathological findings in the bronchial epithelium of pipe and cigar smokers were compared with those found in nonsmokers and cigarette smokers. Pipe and cigar smokers had abnormalities that were intermediate between those of nonsmokers and cigarette smokers, although cigar smokers had pathological changes that in some categories approached the changes seen in cigarette smokers. Others have reported similar findings (144, 233). Several experimental investigations have been conducted to examine the relative tumorigenic activity of tobacco smoke conden- sates obtained from cigarettes, cigars, and pipes. Most of these studies were standardized in an attempt to make the results of the cigar and pipe experiments more directly comparable with cigarette data, and most used the shaved skin of mice for the application of tar. Tar from cigars, pipes, and cigarettes was usually applied on an equal weight basis so that qualitative differences in the tars could be determined. In several experiments, the nicotine was extracted from the pipe and cigar condensates in an attempt to reduce the acute toxic effects that resulted from the high concentration of nicotine frequently found in these products (50, 53, 127, 138, 221, 328). These experimental data suggest that cigar and pipe tobacco condensates have a carcinogenic activity that is comparable to cigarette conden- sates. This is supported by human epidemiologic data for those sites exposed equally to the smoke of cigars, pipes, and cigarettes. The alkaline smoke derived from pipes and cigars is generally not inhaled, and as a result there appears to be a lesser exposure of the lungs and possibly other organs to pipe and cigar smoke than that which occurs due to cigarette smoking. Further, evidence from countries where smokers tend to inhale cigar smoke to a greater degree than smokers do in the United States (I) indicates that rates of lung cancer become elevated to levels approaching those of cigarette smokers. Conclusion 1. Cigarette smoking is the major cause of lung cancer in the United States. 2. Lung cancer mortality increases with increasing dosage of smoke exposure (as measured by the number of cigarettes smoked daily, the duration of smoking, and inhalation pat- terns) and is inversely related to age of initiation. Smokers who consume two or more packs of cigarettes daily have lung cancer mortality rates 15 to 25 times greater than nonsmokers. 62 3. Cigar and pipe smoking are also causal factors for lung cancer. However, the majority of lung cancer mortality in the United States is due to cigarette smoking. 4. Cessation of smoking reduces the risk of lung cancer mortality compared to that of the continuing smoker. Former smokers who have quit 15 or more years have lung cancer mortality rates only slightly above those for nonsmokers (about two times greater). The residual risk of developing lung cancer is directly proportional to overall life -time exposure to cigarette smoke. 5. Filtered lower tar cigarette smokers have a lower lung cancer risk compared to nonfiltered, higher tar cigarette smokers. However, the risk for these smokers is still substantially elevated above the risk of nonsmokers. 6. Since the early 195Os, lung cancer has been the leading cause of cancer death among males in the United States. Among females, the lung cancer death rate is accelerating and will likely surpass that of breast cancer in the 1980s. 7. The economic impact of lung cancer to the nation is consider- able. It is estimated that in 1975, lung cancer cost $3.8 billion in lost earnings, $379.5 million in short-term hospital costs, and $78 million in physician fees. 8. Lung cancer is largely a preventable disease. It is estimated that 85 percent of lung cancer mortality could have been avoided if individuals never took up smoking. Furthermore, substantial reductions in the number of deaths from lung cancer could be achieved if a major portion of the smoking population (particularly young persons) could be persuaded not to smoke. Cancer of the Larynx Introduction Cancer of the larynx was responsible for about 1 percent of cancer deaths in the United States in 1977. It is estimated that in 1982 there will be 10,900 new cases and 3,700 deaths due to this disease (2). Males are affected more commonly than females, but the ratio of new cases and deaths in males and females (now about 6:l) has been narrowing over the last 20 years (240,312). In 1950,1,852 people died of cancer of the larynx. By 1977, this figure had nearly doubled, rising to 3,390. The age-adjusted death rate increased slightly, from 1.1 to 1.2 per 100,000 (Figure 18). There is a considerable difference in this increased death rate when examined by sex and race. Among other than white males, the age-adjusted rate climbed from 1.6'to 3.5 per 100,000 between 1950 and 1977. By contrast, age-adjusted rates of white males rose less, from 2.0 to 2.1. As is seen with lung cancer, mortality rates of females were lower than those of males throughout the study period. Ejetween 1950 and 1977, the age-adjusted mortality rate for white 63 2 +=WHITE HRLES %=UHITE FEHRLES O=NONWHITE IIRLES O II=NCtNHHITE FEllALES 1 ""`0 " 1' `1 1' ' 1" c ' I 1960 1965 1960 1965 1970 1975 CRLENDFIR YEARS females increased from 0.2 to 0.3 per 100,000, while that of other than white females increased from 0.3 to 0.6 per 100,000. Generally, there was a pattern of increasing mortality after middle age (Figures 19 and 20). Among white males 55 years of age or older, mortality rates from cancer of the larynx were higher in 1977 than in 1950. Among other than white males, this pattern was evident for those 35 years of age or older. Both white and other females 45 to 74 years of age had higher mortality rates in 1977 than in 1950. Squamous cell carcinoma is the most common cell type among laryngeal cancers. Approximately 70 percent of the cases involve the glottis and 25 percent involve the supraglottic region. In contrast to lung cancer, the 5-year survival for cancer of the larynx is at present about 60 percent (2). and has been improving over the past 15 years. As a result, the trend over time in death rates from cancer of the larynx is not an accurate reflection of the incidence of this disease. Over the last 30 years, numerous epidemiological, pathological, and experimental investigations have established a strong associa- tion between smoking and cancer of the larynx. One group of scientists (296) conducted a retrospective study of 3,924 patients attending a cancer clinic in Alberta, Canada. The authors estimated that 84 percent of laryngeal cancer among men could be attributed to smoking. Causal Significance of the Association Consistency of the Association More than 25 retrospective studies have examined the relation- ship between smoking and laryngeal cancer. These studies have employed diverse methodology and have been performed in different time periods and in different countries. Regardless of the study design, these studies have found a positive association between smoking and cancer of the larynx. Relative risk ratios for 12 studies up to 1968 (Table 18) were consistently above 2.0. Subsequent studies show similar findings (30,35, &, 52, 113, 114, 134, 142,202,254,296, 299,316,327). The TNCS study (299) and the Hawaiian Study of Five Ethnic Groups (113) have also reported a positive association. Data from studies of populations with low proportions of smokers (e.g., Mormons (165, 266,294) and Seventh Day Adventists (211)) show low laryngeal cancer rates. Six of the major prospective studies have examined the relationship between smoking and laryngeal cancer (Table 19); as in the retrospective studies, a large positive association was consistently noted. 65 fiALES Or 6 30 40 50 60 70 RGE IN YERRS ItJY S-YEAR AGE GROUPS) > I FEnRLES +=1950-1956 *=1957-1963 0 :1964-1970 [3 =1971-1977 10 20 30 40 50 60 76 RGE IN YERRS (BY 5-YEAR AGE CROUPS) 60 Ir FEHALES ? O 1950-1956 % =1967-1963 0 -1964-1970 r3=1971-1977 30 40 50 60 70 80 ROE IN YEARS (6y ~-YERR RGE GROUPSI TABLE lB.-Summary of results of retrospective studies of tobacco use and cancer of the larynx Relative risk rat& Investigator, (reference) all smoke= to nonsmokers Schrek et al., U.S.A. (2441 2.0 Valko, Czechoslovakia (28.2) 3.5 Sadowsky et al., U.S.A. (230) 3.7 Bliimlein, Germany (31) 27.5 Wynder et al., U.S.A. (309) 23.6 Wynder et al., India (309) 3.1 Schwartz et al., France (246) 4.6 Wynder et al., Sweden (3171 6.0 Wynder et al., Cuba (324) (18.9P (males only) DuttaChoudhuri et al., India (77) 4.3 Staxewski. Poland (252) (40.01 hales only) Svobcda, Czechoslovakia (261) 6.3 aComputed awording to the method of J. Cornfield (49). bFiires in parentheses represent ratios based on less than five case nonsmokers. TABLE 19.4Kortality ratios for cancer of' the larynx- prospective studies Study Population size of Cigarette deaths Nonsmokers smokers Comments ACTS S&ate Study 166,000 m&s 24 - All larynx - cancer deaths occurred in smokers British Physicians 34,ooO males 38 1.00 13.00 Includes cancer of larynx and other upper respiratory sites U.S. Veterans 230,mo males 116 1.00 ACS 25-State 356,000 males 67 1.00 Study 483,ooO females 11 100 California males 6smo males 11 - in 9 occupations 11.49 6.52 3.25 Includes buccal. pharyngeal. and laryngeal cancers >2.90 AI1 larynx cancer deaths occurred in smokew Japanese Study 122,OOQ males 38 1.00 13.59 142,800 females 6 1.M) 6.52 = Rat10 derived by comparmg smokers of half a pack with all other smokers 68 TABLE ZO.-Relative risk of laryngeal cancer for males and females bs amount smoked wx day* Number of cigarettes Per Day l-10 11-20 21-m 41+ l-20 21+ Relative NUIllbW Risk Males (N = 243) 16 4.4 a7 13.5 93 17.3 41 34.4 females (N = 48) 19 4.4 23 26.2 Coniidence Limits 1.6 - 12.6 5.3 - 33.1 6.8 - 44.2 12.3 - 96.1 * Risk relative to 1.0 ior nonsmokers. SOURCE: Wynder and Hoffmann (316-l. Strength of the Association In the retrospective studies, the relative risk of laryngeal cancer (Table 18) ranged from 2.0 in a study of 73 U.S. veterans (244) to 40.0 in,a Polish study of 207 males admitted to a chronic disease hospital (252). Two other studies (30, 316) found substantial increases in relative risk among smokers as compared with nonsmokers. Several studies have reported a strong dose-response relationship between the number of cigarettes smoked per day and laryngeal cancer mortality (299, 316). The mortality ratios for male and female cigarette smokers from one of these studies (316) are summarized by daily consumption in Table 20. One study (327) examined the impact of long-term filter cigarette usage on laryngeal cancer risk. After adjustment for duration of smoking, inhalation, and butt length, the relative risk for developing laryngeal cancer was decreased in male and female users of filter cigarettes compared to users of unfiltered cigarettes, although this risk was still substantially greater than that for nonsmokers (Figures 21 and 22). The American Cancer Society 25State Study data (155) also showed a reduced risk of laryngeal cancer among smokers of lower tar and nicotine cigarettes, but this reduction was not statistically significant. In the prospective studies, the mortality ratios for smokers ranged from over 3 among U.S. females to 13 or greater among Japanese males and British male physicians (Table 19). In two of the prospective studies, mortality ratios could not be accurately calculat- ed because all the deaths occurred in smokers, Several of these prospective studies have confirmed the strong dose-response rela- tionship reported in the retrospective studies (Table 21). Specificity of the Association The prospective studies have measured mortality data for a large number of diseases. The specificity of the association is evidenced by 69 CASES N: CONTROLS 10 5 NON F F NF F NF F NF F NF SMOKER l-10 11-20 21-30 3140 41+ 25 126 32 302 16 z FIGURE 21.-Relative risk of developing larynx cancer for males, by number of cigarettes smoked per day and use of filter (F) and nonfilter (NF) cigarettes SOURCE: Wynder (327). the mortality ratios of laryngeal cancer in comparison with other cancers (Appendix Tables A and B). Temporal Relationship of the Association This criterion is supported by the major prospective studies (Table 19) that examined the occurrence of laryngeal cancer in initially healthy groups of smokers and nonsmokers. The temporal relation- ship of the association is strengthened by data from post mortem studies that have evaluated vocal cord histology in groups of smokers and nonsmokers (II, 56, 190, 228). A spectrum of premalignant changes is seen in laryngeal tissue of smokers; this is not found in nonsmokers (see below). 70 45 35 25 15 6 N: CASES CONTROLS B ii v 61 8 NON F NF F NF SMOKER l-20 21+ FIGURE 22.~Relative risk of developing larynx cancer for females, by number of cigarettes smoked per day and use of filter (F) and nonfilter (NF) cigarettes SOURCE: Wynder (327). Coherence of the Association Dose-Response Relationship The finding of a dose-response relationship between smoking and laryngeal cancer incidence and mortality in retrospective and prospective studies strongly supports a causal association. Smoke exposure has been measured by the number of cigarettes smoked per day, the tar and nicotine content of the cigarettes smoked, the depth of inhalation, the number of years smoked, and the age at initiation (269,276), all of which support a direct causal relationship. 71 TABLE 21.-Laryngeal cancer mortality ratios, by amount smoked Population U.S. Veterans Study Japanese Study British Physicians Cigarettes/day Nonsmoker 1-9 l&20 21-39 > 40 Nonsmoker l-19 m-39 4Oi Nonsmoker l-14 15-24 2.5+ Mortality rates 1.00 5.23 9.20 14.78 32.14 ILKI 19.23 27.43 34.13 Male Female 1.00 1.00 5.00 - 7.00 4.00 33.ou 6.50 Comments `Based on less than 20 deaths Includes larynx and other respiratory sites Correlation of Sex Differences in Laryngeal Cancer With Different Smoking Habits Laryngeal cancer is predominantly a disease of males, although the mortality among females has increased over the past 20 years, A male-to-female ratio of 14.9:1 was reported in 1956 (312). The sex ratio decreased to 4.6:1 by 1976. This time trend is consistent with the later adoption of cigarette smoking by females (270) and a possible increase in female alcohol consumption, given the synergy between the two exposures. The greater alcohol consumption among males and the strong association between laryngeal cancer and alcohol consumption (see below) are considered to contribute to the excess of male to female laryngeal cancer mortality. Correlation of Laryngeal Cancer Mortality Among Populations With Different Tobacco Consumption In studies of populations with low proportions of smokers (e.g., Mormons and Seventh Day Adventists), the incidence of laryngeal cancer is substantially lower (79, 165, 166, 211, 294), supporting the causal relationship between smoking and laryngeal cancer. Laryngeal Cancer Mortality and Cessation .of Smoking A few studies have examined the relationship between cigarette smoking cessation and risk for laryngeal cancer. One retrospective study found a marked reduction in risk following cessation among males and females (Figures 23 and 24) and suggested that "10 to 15 years of cessation are required before the long-term smoker's risk approaches. that of a nonsmoker" (327). In the U.S. Veterans and British Physicians studies, ex-smokers had approximately 40 percent 72 N: CASES CONTROLS 16 12 6 4 PRESENT l-3 4-6 7-10 11+ NON SMOKER SMOKER FIGURE 23.-Relative risk of developing larynx cancer for male ex-smokers, by years of smoking tCK?SSation SOURCE: Wynder Wm. of the risk of current smokers for laryngeal cancer; however, the risk was still roughly five times that of the nonsmoker (68,224). Because data were not presented by the number of years off cigarettes, the higher relative risk may be due to higher mortality rates often observed in former smokers (even compared to continuing smokers! during the initial years of smoking cessation. Smoking and Histologic Changes in the Larynx The relationship of smoking habits to precancerous lesions of the larynx was examined in an autopsy series of 148 cases, 24 of whom were nonsmokers (190). Precancerous lesions (dysplasia and carcino- ma in situ) and carcinoma occurred least frequently among non- smokers (4.2 percent). The frequency of these lesions increased from 12.5 percent in light smokers to 22.9 percent in moderate smokers and to 47.2 p¢ in heavy smokers. Similar findings were reported 73 18 6 2 N: CASES 3 CONTROLS 72 SMOKER SMOKER 4-e 7+ NON FIGURE 24.-Relative risk of developing larynx cancer for female ex-smokers, by years of smoking cessation SOURCE. Wynder (3271 from a study of histological changes in the larynx of 942 males aged 21 to 95 (II). These findings lend support to a causal nature of the relationship. Laryngeal Cancer and Non-Cigarette Tobacco Use A few epidemiological studies have examined the relationship between other forms of tobacco use and cancer of the larynx (60, 68, 98, 131). Pipe and cigar smokers develop cancer of the larynx at rates comparable to those of cigarette smokers (i.e., several times those of nonsmokers) (Tables 22 and 23). The similarities of the mortality ratios of cancer of the larynx for smoking of non-cigarette tobacco products suggests that the carcinogenic potentials of smoke from cigars, pipes, and cigarettes are quite similar at this site. The association of smoking of non-cigarette tobacco products to histological changes in the larynx has been examined (II). Among males who smoked cigars and pipes but not cigarettes, only 1 percent 74 TABLE 22.-Mortality ratios for cancer of the larynx in cigar and pipe smokers. A summary of prospective epidemiological studies Study Smoking Type ACS BState Stud? ' British Physicians' ACS 25.State Study U.S. Veterans NOtI- Cigar Pipe Total pipe Cigarette Mixed smoker only Only and cigar only 1.00 5.00 3.50 - 5.06 1.00 - 2.00 1.00 0.60 1.00 - 3.37 36.09 - 1.00 10.33 - 7.28 11.49 `Combmesdata for oral, larynx. and esophagus ' Ratros. rehve in cigarette smokers. 3 Only mortality rata for ages 4.5 to 64 are presented. had no atypical cells and more than 75 percent of the subjects had lesions with 50 to 69 percent atypical cells. Four of the cigar and pipe smokers had carcinoma in situ. Of those who never smoked regularly, 75 percent had no atypical cells. The cigar and pipe smokers had a percentage of cells with atypical nuclei similar to that of cigarette smokers who smoked one to two packs per day. Synergistic Role of Alcohol for Laryngeal Cancer Laryngeal cancer occurs much more frequently in alcoholics than in nonalcoholics (183, 208, 239). Although part of this increased risk for laryngeal cancer among alcohol abusers may be attributed to heavier smoking by this group, there remains a substantial excess risk associated with alcohol use (227). The relative risks of laryngeal cancer by daily consumption of alcohol and cigarettes in 239 male cases and 4,725 controls (Figure 25) suggest a synergy when tobacco usage is combined with chronic alcohol consumption (179). Male smokers of from 11 to 20 and from 21 or more cigarettes per day who consumed 7 ounces or more of alcohol per day had relative risks for laryngeal cancer of 26.8 and 27.2 respectively. The corresponding risks for nondrinking smokers were 6.6 and 12.0. This synergy has also been demonstrated using the Third National Cancer Survey, which suggests that the laryngeal cancer risk for smoking drinkers is approximately 50 percent greater than the sum of the excess risks posed by either behavior alone (85). The mechanism(s) by which these two factors interact is unclear (179,226, 242). Experimental Studies The Syrian golden hamster has been found to be a suitable species for the investigation of cancer of the larynx. The distribution of malignant lesions in the upper airway of the hamster is due not to an unusual susceptibility of the larynx for tumor induction, but rather to the distribution of smoke aerosol precipitation within the upper 75 377-310 0 - 82 - 7