1958 Histologic Types of Lung Carcinoma and Age at Onset Michaela Kreuzer, Ph.D.1,2 Lothar Kreienbrock, Ph.D.1 Klaus M. Müller, M.D.3 Michael Gerken, M.D.1 Erich Wichmann, M.D., Ph.D.1 1 GSF-National Research Center for Environment and Health, Institute of Epidemiology, Neuherberg, Germany. 2 Federal Office of Radiation Protection, Institute of Radiation Hygiene, Neuherberg, Germany. 3 Berufsgenossenschaftliche Kliniken Bergmannsheil/Ruhr-Universität Bochum, Institute of Pathology, Bochum, Germany. BACKGROUND. Previous research has demonstrated that adenocarcinoma is the leading cell type among patients with early age onset lung carcinoma. An increase in adenocarcinoma at the expense of squamous cell carcinoma in general was observed in recent years and may be due to the smoking of filtered cigarettes. METHODS. To rule out whether shifts in smoking patterns or other etiologic factors are responsible for the high rates of adenocarcinoma in young patients, personal interviews regarding smoking, occupation, and family history of cancer were conducted in 251 young patients (age # 45 years) and 2009 older patients (ages 55– 69 years) with histologically confirmed lung carcinoma from selected study clinics in Germany between 1990 and 1996. RESULTS. Young male patients were found to have significantly more adenocarcinomas (41%) than older male patients (28%), whereas adenocarcinomas were dominant in young and older women (43% and 47%, respectively). Because smoking patterns were different between young and older patients, the authors stratified for comparable levels of smoking exposure. Histology did not differ in never smokers (dominance of adenocarcinomas in both age groups) and in male heavy smokers (dominance of squamous cell carcinomas in both age groups), whereas young male low dose smokers showed significantly more cases of adenocarcinoma than older low dose smokers. A family history of lung carcinoma was significantly higher in young patients compared with older patients, but no association with histologic type was observed. CONCLUSIONS. The results of the current study show that differences in the histologic type of lung carcinoma based on age at onset can be explained in part by differences in smoking patterns. However, there still are unknown factors that appear to favor the development of adenocarcinoma in the young. Cancer 1999;85: 1958 – 65. © 1999 American Cancer Society. KEYWORDS: lung carcinoma, smoking, histology, family. C Supported by the Federal Office of Radiation Protection, Salzgitter, Germany (grant St Sch 1066, 4074, 4074/1, 4006, 4112). Address for reprints: Dr. Michaela Kreuzer, Federal Office of Radiation Protection, Institute of Radiation Hygiene, Ingolstaedter Landstrasse 1, D-85764 Neuherberg, Germany. Received May 22, 1998; revisions received September 21, 1998 and January 11, 1999; accepted January 11, 1999. © 1999 American Cancer Society arcinoma of the lung occurs most commonly in the sixth and seventh decades of life and is rarely found in young patients. Most studies over the past decades reported adenocarcinoma as the most frequent cell subtype among patients with early onset lung carcinoma of both genders,1–12 which is in contrast to squamous cell carcinoma as the most frequent type in older men.13 The predominance of adenocarcinoma in the young could reflect either a propensity of young individuals to develop this subtype or an increase in this subtype in recent years, which initially is reflected in younger cohorts. Some studies suggest that the reduction of tar yield and the introduction of filtered cigarettes in the 1960s could have favored the development of adenocarcinoma at the expense of squamous cell carcinoma.14,15 Such shifts in the incidences of different histologic types of lung carcinoma have been found in the United States, where adenocarcinoma had become the most frequent subtype at the beginning of the 1980s.16 –18 In Europe, squamous cell carcinoma still predomi- Lung Carcinoma Histology and Age at Onset/Kreuzer et al. 1959 FIGURE 1. Distribution of histologic types by age group (in years) at diagnosis (n 5 4500). SCLC: small cell lung carcinoma; AC: adenocarcinoma; SqCC: squamous cell carcinoma. nates in most countries. Levi and colleagues19 reported a remarkable rise in adenocarcinoma incidence in men and women from Switzerland, with rates of adenocarcinoma incidence among young adults in the early 1990s being more than three times higher than rates of squamous cell carcinoma. Up to now, it is still inconclusive whether or not high rates of adenocarcinoma in young subjects are mainly attributable to shifts in smoking pattern or whether other still unknown factors are causal. In a previous analysis of a case– control study of lung carcinoma, we found smoking and a possible genetic predisposition as risk factors for lung carcinoma in young subjects (#45 years).20 In addition, a high percentage of adenocarcinomas in this young group compared with older patients (55– 69 years) was observed. The aim of the present analysis of these data is to further investigate the association between histologic type and age of onset of lung carcinoma. This study provides a large number of young and older lung carcinoma patients of both genders, a detailed quantification of life-long smoking and occupational exposure, and information on family history of cancer. PATIENTS AND METHODS Data of patients were derived from a case– control study on lung carcinoma conducted between 1990 and 1996 in several regions in East Germany (Thuringia and Saxony) and West Germany (Eastern Bavaria, the Saarland, parts of Northrhine-Westfalia, and Rhineland Palatinate).21–23 Patients with histolog- ically or cytologically confirmed lung carcinoma with primary tumors were recruited from 15 study clinics in this study region. Patients were eligible if 1) they were ,75 years of age, 2) they were resident in the study region, 3) they had lived in Germany for .25 years, 4) the interviews were conducted within 3 months of diagnosis, 5) they were not too ill, and 6) their tumor type was not “carcinoid.” The response rate among eligible patients was 76%. According to World Health Organization (WHO) classification, we coded the microscopic noticeably leading histologic type into four groups: small cell lung carcinoma (SCLC), squamous cell carcinomas (SqCC), and adenocarcinomas (AC). Due to small numbers we combined large cell carcinomas, mixed types, and tumors, for which no classification was possible, into the category “other.” About 60% of diagnoses were derived from histology, 20% were derived from cytology, and 20% were derived from both. Diagnostic and cell type procedures were established by each participating hospital. We used a diagnosis from cytology only if a diagnosis from histology was not present. A reference pathologist reviewed about 70% of the pathologic material without any knowledge of age, gender, or tumor type diagnosed by the clinical pathologist. To avoid misclassification due to different pathologists, diagnoses of tumor histology from the reference pathologist were used where available, with missing reference histology replaced by a diagnosis from the clinical pathologist. A total of 4500 lung carcinoma cases were included in the study. Figure 1 1960 CANCER May 1, 1999 / Volume 85 / Number 9 shows the frequencies of cell subtypes according to age at diagnosis in 5-year categories. A clear association between age and histology was observed. The highest frequencies of adenocarcinomas were seen in the younger age groups, whereas squamous cell carcinomas were most frequent in the older age groups. We classified a young and older age group according to our previous analysis.20 Patients were defined as young if they were #45 years of age and older if they were 55– 69 years of age. Subjects between 46 –54 years of age were excluded to clearly separate the two age groups, and those .69 years of age were excluded, because their recollection of past exposures might be less accurate than that in the younger age groups. After the patient’s informed consent was obtained, an interview with the patient was conducted by trained interviewers at the beside or at home if the patient had already been discharged. A standardized questionnaire was used to determine basic demographic characteristics in addition to details on active smoking history, occupational history, and a family history of cancer. Subjects were defined as smokers if they had ever smoked regularly (at least 1 cigarette per day, 4 cigarillos per week, or 3 cigars or 3 pipes per week) for at least 6 months. A life-long history of smoking exposure was explored in a concept of smoking periods. In each period, information was available on the type of tobacco, filter usage, daily consumption, duration in years, times of cessation, and year of starting. All occupations during a patient’s lifetime were evaluated in a concept of job periods. Subjects were classified as having been exposed occupationally to known or suspected lung carcinogens if they had worked for at least 6 months in a job entailing exposure to recognized (A-list) or suspected (B-list) carcinogens, mostly based on the International Agency for Research on Cancer monographs program.24 Information on history of cancer among first-degree relatives was gathered, including age at diagnosis and cancer site. Subjects were defined as having cancer in the family if at least one parent or sibling with cancer was reported. This factor was defined for lung carcinoma, carcinoma at age #45 years, and lung carcinoma at age #45 years. One-year measurements of radon concentrations in the bedroom and living room of the patient’s last dwelling were obtained by a-track detectors.25 To date, these measurements have been completed for '70% of all patients. For this subsample, the time-weighted average of radon concentrations measured in the bedroom and living room was calculated. Variables of interest were analyzed by means of cross tabulations on contingency tables. Chi-square statistics with associated significance levels were calculated. Tests for differences in histology by age were performed after stratification by gender and, separately, for different exposure levels of smoking. RESULTS Table 1 shows characteristics of the study population. A total of 251 young and 2009 older lung carcinoma cases were included in the analysis. The ratio between men and women was appreciably lower in the young group (2.7:1.0) than in the older group (5.6:1.0). The relative frequency of adenocarcinoma was substantially higher in young men (41%) than in older men (28%), whereas squamous cell carcinoma was more common among older men (42%) than among young men (26%). Both age groups of women showed adenocarcinoma as the leading tumor type (43% in the young women, 47% in the older women). Significantly more young female patients were smokers (91%) compared with older female patients (68%), whereas there was no statistically significant difference in the smoking status between young and older men (97% and 99%, respectively). On average, regardless of gender, older patients smoked about 15 years longer than young patients and smoked about four cigarettes fewer per day. Life-long consumption of filtered cigarettes was considerably higher in young patients (62% of men and 92% of women) than in older patients (13% of men and 47% of women). A family history of lung carcinoma in first-degree relatives revealed a significantly increased proportion in young men compared with older men (P 5 0.0045). Young patients reported significantly more cancer with age at diagnosis , 46 years in first-degree relatives than older patients (P 5 0.000 among men; P 5 0.004 among women). The most notable difference was a higher proportion of familial aggregation of early onset lung carcinoma among young lung carcinoma patients compared with older patients (2.2% among young male cases vs. 0.1% among older male cases; 2.9% vs. 1.0% among women, respectively). No significant differences in exposure to known or suspected occupational carcinogenic substances were observed between age groups in both genders. Also, in the indoor air radon concentrations did not differ between young and older patients. The median radon concentration was between 40 Bq/m3 and 42 Bq/m3 among young and older men and among older women. It was somewhat smaller among young women, but the sample size was very small in this group. Because of this lack of a statistically significant difference between age groups, occupational exposure and radon exposure were not controlled for in the analysis. Lung Carcinoma Histology and Age at Onset/Kreuzer et al. 1961 TABLE 1 Characteristics of Lung Carcinoma Patients by Gender and Age Group Men Women Characteristic <45 yrs 55–69 yrs <45 yrs 55–69 yrs Total Mean age (yrs) Histology (%) Small cell carcinoma Squamous cell carcinoma Adenocarcinoma Other Smoking status (%) Never smokersa Ever smoked cigarettes Other smokers (pipes, cigars) Characteristics of cigarette smokers Duration of smoking (yrs)b Average daily consumption (cigarettes/day) Life-long filtered cigarettes only (%) Occupation (%) Ever A-listc Ever A or B-listd Family history of cancer (at least one parent or sibling) (%) Lung cancer Cancer at age #45 yrs Lung cancer at age #45 yrs Radon concentration at home Number of subjects with valid 1-year measurements Median (Bq/m3) 183 41 1709 63 68 42 300 62 24.0 26.2 41.0 8.7 22.5 42.1 28.0 7.5 30.9 19.1 42.7 7.4 20.3 22.7 47.0 10.0 3.3 96.7 0.0 1.3 97.2 1.5 8.8 91.2 0.0 31.7 67.3 1.0 22.9 6 4.3 22.8 6 9.3 62 37.3 6 10.7 19.5 6 9.2 13 20.9 6 6.5 19.2 6 12.0 92 35.5 6 10.9 15.2 6 8.5 47 15.8 38.8 16.4 43.2 2.9 16.2 4.0 17.0 10.3 8.6 2.2 6.3 3.2 0.1 9.0 11.9 2.9 10.9 3.4 1.0 93 41.1 1207 41.5 36 35.4 182 39.7 a Never smoked more than one cigarette per day regularly for longer than 6 months. Values represent mean 6 standard deviation. c Ever exposed to a known lung carcinogen (A-list). d Ever exposed to a known or suspected lung carcinogen (A- or B-list). b Smoking and Histology Table 2 provides the distribution of histologic type by smoking exposure among men. Never smokers, as expected, exhibited the highest frequency of adenocarcinoma in both male age groups (67% in young men and 57% in older men). Among young male cigarette smokers, statistically significantly more adenocarcinomas and fewer squamous cell carcinomas were observed than among older men. After stratification of tobacco consumption into three categories, a strong decrease in the frequency of adenocarcinoma and small cell lung carcinoma accompanied by an increase in squamous cell carcinomas with increasing average daily consumption was observed in the young group. This tendency was not found or was found only in a very weak form in the older group. Among heavy smokers (.30 cigarettes per day), the histologic type between older and young cases did not differ statistically, but it is notable that differences in histology by age group were seen in low and medium dose smokers. The group of weak smokers (,14 cigarettes per day) showed significantly more adenocarcinomas (46%) and fewer squamous cell carcinomas (13%) in the young group compared with the older group (27% and 41%, respectively). Because the underlying duration of smoking between both age groups was completely different, we divided consumption of tobacco into a group of subjects who smoked for ,30 years and those who smoked .30 years. Again, in weak smokers (,30 years, ,20 cigarettes/day), we found significantly more adenocarcinomas (48%) and fewer squamous cell carcinomas (25%) in young men than in older men (31%, 40%, respectively) and found no differences in histologic type between young and older “heavy” cigarette smokers. An appreciably higher proportion of squamous cell carcinomas (37%) among young smokers of nonfiltered cigarettes was seen compared with pure filtered cigarette smokers (20%) (Table 2). This effect was not so pronounced in older patients. Again “heavy smokers” (life-long nonfiltered or mixed filtered cigarettes) exhibited no statistical difference in tumor type 1962 CANCER May 1, 1999 / Volume 85 / Number 9 TABLE 2 Distribution of Histologic Type by Age Group and Various Smoking Variables among Men Only Never Smokers and Cigarette Smokers (Pure Smokers of Pipes, Cigars, or Cigarillos Excluded) <45 yrs: cell type (%) 55–69 yrs: cell type (%) Smoking variable SCLC SqCC AC Other Total SCLC SqCC AC Other Total P valuea Never smokers Ever smokers Daily consumption 1–14 cig/day 15–29 cig/day 301 cig/day Duration/consumption ,30 yrs, ,20 cig ,30 yrs, 201 cig 301 yrs, ,20 cig 301 yrs, 201 cig Filter usage Filtered cig only Nonfiltered and mixed filtered cig — 25.4 33.3 26.6 66.7 41.2 — 6.8 6 177 13.0 22.1 13.0 41.8 56.5 27.6 17.4 8.5 23 1653 — 0.000 37.5 24.8 18.8 12.5 27.7 34.4 45.8 41.3 37.5 4.2 6.6 9.4 24 121 32 23.7 22.2 17.2 40.5 41.8 44.6 27.3 27.3 30.9 8.6 8.7 7.3 477 972 204 0.02 0.003 0.74 22.5 27.8 0 25.0 25.4 25.8 100.0 37.5 47.9 38.1 0 25.0 4.2 8.3 0 12.5 71 97 1 8 20.0 19.8 25.5 18.2 39.5 38.5 41.0 44.2 31.2 31.3 25.1 29.5 9.3 10.4 8.3 8.2 215 96 792 550 0.03 0.20 — — 29.4 19.1 20.2 36.8 44.0 36.8 6.4 7.4 109 68 22.2 22.0 39.2 42.1 30.2 27.4 8.5 8.5 212 1439 0.003 0.42 SCLC: small cell lung carcinoma; SqCC: squamous cell carcinoma; AC: adenocarcinoma; cig: cigarettes. a P values were determined by chi-square test for differences in histologic types between age groups (each row indicates one test). TABLE 3 Distribution of Histologic Type by Age Group and Various Smoking Variables among Women Only Never Smokers and Cigarette Smokers (Pure Smokers of Pipes, Cigars, or Cigarillos Excluded) <45 yrs: cell type (%) 55–69 yrs: cell type (%) Smoking variables SCLC SqCC AC Other Total SCLC SqCC AC Other Total P valuea Never smokers Ever smokers Daily consumption 1–14 cig/day 15–29 cig/day 301 cig/day Duration/consumption ,30 yrs, ,20 cig ,30 yrs, 201 cig 301 yrs, ,20 cig 301 yrs, 201 cig Filter usage Filtered cig only Nonfiltered and mixed filtered 16.7 32.3 33.3 17.7 16.7 45.2 33.3 4.8 6 62 10.5 24.4 11.6 26.8 66.3 38.5 11.6 10.2 95 202 — 0.18 40.0 32.1 11.1 12.0 21.4 22.2 40.0 46.4 55.6 8.0 0 11.1 25 28 9 24.0 25.3 27.3 29.0 25.3 27.3 37.0 38.5 45.5 10.0 11.0 0 100 91 11 0.09 0.10 0.49 41.9 11.8 0 0 11.6 35.3 0 0 41.9 52.9 0 50.0 4.6 0 0 50.0 43 17 0 2 26.2 66.7 21.9 29.0 19.1 33.3 29.4 29.0 47.6 0 37.8 31.6 7.1 0 10.9 10.5 42 3 119 38 0.44 — — — 29.8 60.0 17.5 20.0 47.4 20.0 5.3 0 57 5 26.0 23.8 26.0 28.6 36.5 39.1 11.5 8.6 96 105 0.28 — SCLC: small cell lung carcinoma; SqCC: squamous cell carcinoma; AC: adenocarcinoma; cig: cigarettes. a P values were determined by chi-square test for differences in histologic types between age groups (each row indicates one test). between young and older patients, whereas life-long smokers of filtered cigarettes differed according to histologic type. In the large group of never smoking older women (n 5 95), adenocarcinomas were dominant (66%). Only 6 young female lung carcinoma patients had never smoked in their life, among whom 1 case of adenocarcinoma occurred. The proportion of adenocarcinomas in female cigarette smokers (Table 3) was reduced compared with never smokers but remained the dominant subtype in both age groups (45% in the young group and 39% in the older group), followed by small cell lung carcinoma. No statistically significant differences in histologic type by age group were observed, even after stratification for comparable levels of exposure. The detailed analyses of subgroups of smoking women are difficult to evaluate due to the very small numbers. Lung Carcinoma Histology and Age at Onset/Kreuzer et al. 1963 TABLE 4 Distribution of Histologic Type by Age Group and Family History of Cancer (at Least One Parent or Sibling) <45 yrs: cell type (%) Family history Lung carcinoma Yes No Cancer at age #45 yrs Yes No 55–69 yrs: cell type (%) SCLC SqCC AC Other Total SCLC SqCC AC Other Total 33.3 24.7 29.2 23.4 29.2 42.7 8.3 9.2 24 218 20.1 22.4 38.9 39.1 33.1 30.6 7.9 7.9 139 1845 43.5 23.6 26.1 25.0 26.0 42.7 4.4 8.6 23 220 12.3 22.5 41.5 38.9 29.2 30.7 16.9 7.9 65 1923 SCLC: small cell lung carcinoma; SqCC: squamous cell carcinoma; AC: adenocarcinoma. Family History of Cancer and Histology There was no evidence of differences in histology between patients with or without a family history of lung carcinoma in either in the young group or in the older age group (Table 4). The same was seen for a family history of early onset carcinoma. It is notable that there was no tendency for adenocarcinoma to be associated with a reported family history of lung carcinoma. Only one young patient who had reported a family history of lung carcinoma was a nonsmoker. A stratification of groups into weak smokers and heavy smokers did not change the distribution of cell subtypes by reported family cancer history. DISCUSSION Two findings have emerged from our analysis of histologic type and age of onset of lung carcinoma. The first was that there are many similarities between the findings in our study population and others reported in literature. In all of the reports that include both men and women, there are more women in the younger group than in the general population of patients with lung carcinoma.7,10,12,26,27 We found a male-to-female ratio of 2.7:1 in the young group and 5.6:1 in the older group. The relative increase in the number of women among the younger patients may reflect the more recent increase in smoking among women. Previous analysis of population controls matched by age, gender, and region to our patients with lung carcinoma exhibited 52% current or exsmokers among young female controls and 36% among older female controls.20 Several studies conducted over the past three decades reported a preponderance of adenocarcinoma in young patients.1–12,28 This is consistent with our results of about 41% adenocarcinoma in young male patients (43% in young female patients) compared with 28% adenocarcinoma in older male patients (47% in older female patients). Only a few of these studies included data on smoking history,1,10 and none reported histologic type analyses stratified by smoking exposure. The most important new finding in our study, however, was that differences in histology by age could be explained in part by differences in smoking pattern and gender. It is evident that the proportion of smokers in young women is higher than in older women, that young patients smoke more filtered cigarettes and more cigarettes per day and are less likely to be long term smokers than older patients, regardless of gender. Among the group of women, no statistically significant differences in age-related histology were observed, but the statistical power to detect such differences was small. It should be noted that there is a markedly greater number of small cell carcinomas among young women compared with older women. This either could be random (due to the small numbers) or may reflect the trend of a reduction in adenocarcinomas and an increase in small cell carcinomas in female smokers, which was found in the United States.29 Histology of men differed significantly by age. However, after stratification for comparable smoking exposure levels, some of the differences disappeared. For example, adenocarcinoma was the most frequent histologic type in life-long never smokers in both age groups. Among “heavy male smokers,” like those who smoked more than 30 cigarettes per day on average, those who were life-long smokers of nonfiltered or mixed filtered cigarettes, and those who smoked more than 20 cigarettes longer than 30 years, squamous cell carcinoma was the most predominant cell subtype in young and older male patients, and no statistically significant differences were observed between the age groups. Differences in histology by age remain only in low and medium dose male smokers, with a higher frequency of adenocarcinomas in young patients than in older patients. Most recent studies suggest that the increasing 1964 CANCER May 1, 1999 / Volume 85 / Number 9 predominance of adenocarcinoma over squamous cell carcinoma may be due in part to the reduced risk of squamous cell carcinoma associated with life-long filtered cigarette smoking.14,30 Filters remove larger particles in cigarette smoke, thus reducing the deposition of those particles into central airways, where the squamous cell carcinoma develops preferentially.31,32 This could lead to a reduction in the incidence of squamous cell carcinoma but not of adenocarcinoma, which occurs primarily in peripheral lung areas.15,33 In fact, in the group of young men, we observed more frequent squamous cell carcinomas among nonfiltered and mixed filtered cigarette smokers (37%) than among life-long smokers of filtered cigarettes (20% squamous cell carcinoma). Among older men, this effect was not so clear (42% vs. 39%). It is striking that histology in nonfiltered and mixed filtered cigarette smokers did not differ between young and older men, whereas significant differences exist in life-long filtered cigarette smokers. Several factors, however, are likely to have contributed to a higher percentage of adenocarcinoma in the young filtered cigarette smokers compared with the older filtered cigarette smokers. First, the cumulative consumption of filtered cigarettes over a whole life time is much greater in older men than in young men, because they have smoked about 15 years longer on the average. It is possible that the necessary inciting stimulus for the transformation of the respiratory mucosa to squamous cell carcinoma must act over a longer time span to be effective.1 In addition, tar and nicotine contents have been reduced strongly in Germany since the 1960s. Because we have no information on cumulative tar and nicotine exposure among young and older men, a possible interaction with age and its effects on histology could not be determined. On the whole, young mild smokers are likely to have less cumulative tobacco exposure than older mild smokers and are therefore closer to the nonsmokers, who are known for a predominance of adenocarcinomas for reasons other than smoking. Environmental exposure to other agents or genetic factors also may play a role in differences in histology by age. It is widely assumed that cancers diagnosed at an earlier age may indicate a larger role of genetic predisposition. Our previous analysis20 showed a threefold increased lung carcinoma risk in the young group (#45 years) if at least one lung carcinoma case was reported in a first-degree relative and a twofold increased lung carcinoma risk if cancer at a young age in a first-degree relative was reported. No such influence was seen in the older group (55– 69 years). A possible genetic predisposition also is supported strongly by the segregation analysis of Sell- ers,34,35 who suggested that Mendelian inheritance of a rare major autosomal gene may produce lung carcinoma at an earlier age of onset. Because adenocarcinoma is seen more often at an earlier age than other cell types, it is surprising that the excess of reported family history of cancer is not observed among patients with adenocarcinoma. This phenomenon is supported by data presented by Ambrosone et al.36 They observed that, of all histologic types of lung carcinoma, squamous cell carcinoma is the type most associated with familial clustering of malignancy, and adenocarcinoma is the least. Sellers et al.37 found little direct evidence of increased genetic predisposition among patients with particular types of lung carcinoma. A strength of this study is that all patients were interviewed in person; no data were obtained from next of kin or other surrogates. Use of closely supervised, trained interviewers and standardized questionnaires also served to increase confidence in the results. Refusal rates have consistently been under 23% in the young patients and under 27% in the older patients. No statistical differences in histologic type were observed between study participants and refusals, so that little selection bias is believed to exist. Review of a subsample ('70%) of all pathologic material exhibited a strong association between histologic diagnosis of the clinical pathologist and the reference pathologist. Eighty-six percent of young patients with lung carcinoma diagnosed as small cell carcinoma by the clinical pathologist had their cell subtype confirmed by the reference pathologist (82% in older patients), 94% of young patients with adenocarcinoma had their cell subtype confirmed by the reference pathologist (80% in older patients), and 62% of young patients with squamous cell carcinoma had their cell subtype confirmed by the reference pathologist (68% in older patients). A possible classification bias due to different clinical pathologists was minimized by using the diagnoses of the reference pathologist, where available. CONCLUSIONS Adenocarcinoma is the leading cell subtype in young male and female patients with lung carcinoma, whereas, among older male patients, squamous cell carcinoma predominates. Differences in histologic type between patients with onset of lung carcinoma at an early age and older patients can be explained in part by differences in smoking patterns. After stratification for comparable levels of smoking exposure, the groups of heavy smokers and never smokers exhibited no statistically significant differences in cell subtype compared with low dose smokers, like life-long smok- Lung Carcinoma Histology and Age at Onset/Kreuzer et al. ers of filtered cigarettes. Results indicate a genetic predisposition in young patients that shows no influence on histology. There still remain unknown factors that appear to favor the development of adenocarcinoma in young patients with lung carcinoma. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Kyriakos M, Webber B. Cancer of the lung in young men. J Thorac Cardiovasc Surg 1974;67:634 – 48. Putnam JS. Lung carcinoma in young adults. JAMA 1977;238: 35– 6. DeCaro L, Benfield JR. Lung cancer in young persons. J Thorac Cardiovasc Surg 1982;83:372– 6. Pemberton HJ, Nagorney MD, Gilmore CJ, Taylor FW, Bernartz EP. Bronchogenic carcinoma in patients under 40 years. Ann Thorac Surg 1985;36:509 –15. Larrieu AJ, Jamieson WR, Nelems JM, Fowler R, Yamamoto B, Leriche J, et al. Carcinoma of the lung in patients under 40 years of age. Am J Surg 1985;149:602–5. Tsugane S, Watanabe S, Sugimura H, Arimoto H, Shimosato Y, Suemasu K. Smoking, occupation and family history in lung cancer patients under fifty years of age. Jpn J Clin Oncol 1987;17:309 –17. Jubelirer SJ, Wilson RA. Lung cancer in patients younger than 40 years of age. Cancer 1991;67:1436 – 8. Icard P, Regnard JF, de Napoli S, Rojas Miranda A, Dartevelle P, Levasseur P. Primary lung cancer in young patients: a study of 82 surgically treated patients. Ann Thorac Surg 1992;54:99 –103. Bourke W, Milstein D, Giura R, Donghi M, Luisetti M, Rubin AH, et al. Lung cancer in young adults. Chest 1992;102:1723–9. Green LS, Fortoul TI, Ponciano G, Robles C, Rivero O. Bronchogenic cancer in patients under 40 years old. The experience of a Latin American country. Chest 1993;104:1477– 81. Rocha MP, Fraire AE, Guntupalli KK, Greenberg SD. Lung cancer in the young. Cancer Detect Prev 1994;18:349 –55. Antkowiak JG, Regal AM, Takita H. Bronchogenic carcinoma in patients under age 40. Ann Thorac Surg 1989;47:391–3. Wichmann H-E, Molik B, Jöckel K-H, Jahn I, Müller K-M. Ergebnisse zur Histologie des Bronchialkarzinoms aus einer epidemiologischen Untersuchung in Nordrhein-Westfalen und Norddeutschland. Pneumologie 1990;44:1251– 8. Stellman SD, Muscat JE, Thompson S, Hoffmann D, Wynder EL. Risk of squamous cell carcinoma and adenocarcinoma of the lung in relation to lifetime filter cigarette smoking. Cancer 1997;80:382– 8. Charloux A, Quoix E, Wolkove N, Small D, Pauli G, Kreisman H. The increasing incidence of lung adenocarcinoma: reality or artefact? A review of the epidemiology of lung adenocarcinoma. Int J Epidemiol 1997;26:14 –23. Dodds L, Davis S, Polissar L. A population-based study of lung cancer incidence by histological type. J Natl Cancer Inst 1986;76:21–9. Thun MJ, Lally CA, Flannery JT, Calle EE, Flanders D, Heath CW. Cigarette smoking and changes in the histopathology of lung cancer. J Natl Cancer Inst 1997;89:1580 – 6. Travis WD, Travis LB, Devesa SS. Lung cancer. Cancer 1995; 75:191–202. Levi F, Franceschi S, La Vecchia C, Randimbison L, Te V. Lung carcinoma trends by histologic type in Vaud and Neuchatel, Switzerland, 1974 –1994. Cancer 1997;79:907–14. 1965 20. Kreuzer M, Kreienbrock L, Gerken M, Heinrich J, BrüskeHohlfeld I, Müller K-M, et al. Risk factors for lung cancer in young adults. Am J Epidemiol 1998;147:1028 –37. 21. Kreienbrock L, Wichmann H-E, Gerken M, Heinrich J, Goetze H-J, Kreuzer M, et al. The German Radon Project Feasibility of methods and first results. Rad Prot Dos 1992; 45:643–9. 22. Wichmann H-E, Kreienbrock L, Kreuzer M, Gerken M, Dingerkus G, Wellmann J, et al. Lungenkrebsrisiko durch Radon in der Bundesrepublik Deutschland (West). In: Wichmann HE, Schlipköter HW, Fülgraff G, editors. Fortschritte in der Umweltmedizin. Landsberg/Lech: Ecomed Verlag, 1998. 23. Jöckel K-H, Brüske-Hohlfeld I, Wichmann HE Lungenkrebsrisiko durch berufliche Expositionen. In: Wichmann HE, Jöckel KH, Robra BP, editors. Fortschritte in der Epidemiologie. Landsberg/Lech: Ecomed Verlag, 1998. 24. Boffetta P, Kogevinas M, Simonato L, Wilbourn J, Sarracci R. Current perspectives on occupational cancer risks. Int J Occup Environ Health 1995;1:315–25. 25. Keller G, Kappel RJA, Gerken M, Wellmann J, Kreuzer M, Kreienbrock L, et al. Initial results of indoor radon measurements within the German radon studies. Environ Int 1996; 22:S665–70. 26. Ganz PA, Vernon SE, Preston D, Coulson WF. Lung cancer in younger patients. West J Med 1980;133:373– 8. 27. Kennedy A. Lung cancer in young adults. Br J Dis Chest 1972;66:147–54. 28. Cha Q, Chen Y, Du Y. The trends in histological types of lung cancer during 1980 –1988, Guangzhou, China. Lung Cancer 1997;17:219 –30. 29. Yesner R. Lung cancer, pathogenesis and pathology. Clin Chest Med 1993;14:17–23. 30. Stellman SD, Muscat JE, Hoffmann D, Wynder EL. Impact of filter cigarette smoking on lung cancer histology. Prev Med 1997;26:451– 6. 31. Zheng T, Holford TR, Boyle P, Chen Y, Ward BA, Flannery J, et al. Time trend and the age-period-cohort effect on the incidence of histologic types of lung cancer in Connecticut, 1960 –1989. Cancer 1994;74:1556 – 67. 32. Wynder EL, Hoffmann D. Smoking and lung cancer: scientific challenges and opportunities. Cancer Res 1994; 54:5284-95. 33. Morabia A, Wynder EL. Cigarette smoking and lung cancer cell types. Cancer 1991;68:2074 – 8. 34. Sellers TA, Bailey Wilson JE, Elston RC, Wilson AF, Elston GZ, Ooi WL, et al. Evidence for Mendelian inheritance in the pathogenesis of lung cancer. J Natl Cancer Inst 1990;82: 1272–9. 35. Sellers TA, Potter JD, Bailey Wilson JE, Rich SS, Rothschild H, Elston RC. Lung cancer detection and prevention: evidence for an interaction between smoking and genetic predisposition. Cancer Res 1992;52:2694s–7s. 36. Ambrosone CB, Rao U, Michalek AM, Cummings KM, Mettlin CJ. Lung cancer histologic types and family history of cancer. Analysis of histologic subtypes of 872 patients with primary lung cancer. Cancer 1993;72:1192– 8. 37. Sellers TA, Elston RC, Atwood LD, Rothschild H. Lung cancer histologic type and family history of cancer. Cancer 1992;69:86 –91.