Epidemiology of Bronchioloalveolar Carcinoma1characteristics of lung adenocarcinoma. Received 10/24/91. ‘ Supported by Contract N01-CP-91023 from the National Cancer Institute. To

Vol. 1, 3(9- 344, July/August 1992 Cancer Epidemiology, Biomarkers & Prevention 339

Epidemiology of Bronchioloalveolar Carcinoma1

Roni T. Falk,2 Linda W. Pickle,3 Elizabeth T. H. Fontham,S. Donald Greenberg, Helen L. Jacobs, Pelayo Correa,and Joseph F. Fraumeni, Jr.

Epidemiology and Biostatistics Program, Division of Cancer Etiology,National Cancer Institute, Bethesda, Maryland 20892 [R. T. F., L. W. P.,

I. F. F.]: Pathology Department, Louisiana State University Medical

Center, New �I�rleans, Louisiana 701 12 [E. T. H. F., P. C.]; Pathology

Department, Baylor College of Medicine, Houston, Texas 77030[S. D. G.]; and 1816 New Hampshire Avenue NW, Washington, DC

20009 [H. L. I.]

Abstract

Descriptive features of bronchioloalveolar carcinoma(BAC) are presented using Surveillance, Epidemiology

and End Results Program population-based incidencedata from 1973 through 1987, along with risk factors

from histologically confirmed cases of BAC identified

in a hospital-based case-control study conducted inLouisiana between 1979 and 1982. Compared to the

rising incidence of lung cancer overall, BAC rates haveremained relatively constant, accounting for less than3% of all lung cancer. BAC incidence rates were higherin males, yet it explained proportionately more of the

total lung cancer incidence in females. In the case-control study, 21 of the 33 cases originally ascertainedfrom hospital pathology records were histologicallyconfirmed as BAC. Most cases smoked cigarettes, witha 4-fold risk for ever smoking. Risks tended to increasewith smoking intensity (reaching 10-fold for more than1.5 packs/day) and duration (reaching 5-fold for morethan 45 years of smoking). Following 10 or more yearsof employment, there was a 4-fold risk associated withmotor freight occupations, along with nonsignificantexcesses among construction workers, petroleummanufacturers, and sugar cane farmers. Cases weremore likely than controls to have had emphysema or tohave had a close family member with lung cancer.Although based on small numbers, this study suggeststhat BAC shares many of the epidemiologicalcharacteristics of lung adenocarcinoma.

Received 10/24/91.

‘ Supported by Contract N01-CP-91023 from the National Cancer

Institute.To whoni requests for reprints should be addressed, at Epidemiology

and Biostatistics Program. Division of Cancer Etiology, National Cancer

Institute, Executive Plaza North, Room 430C, Bethesda, MD 20892.

Present address: National Center for Health Statistics, Hyattsville, MD

20782.

IntroductionBAC4 is a relatively rare lung tumor, accounting for lessthan 3% of incident lung cancers in the United States (1).Diagnostic controversy has surrounded BAC, in part be-cause its histogenesis is uncertain, it is difficult to distin-guish from adenocarcinoma, and the question of multi-centric versus unicentric origin remains open (2). BACappears to be a heterogeneous group of tumors that maybe derived from type II pneumocytes, Clara cells, orbronchial mucus cells resembling mucus-secreting ad-enocarcinomas (2-7). Little is known about its etiology,with clues being generated mainly from clinical series (3-9). Earlier reports have suggested that in relation to otherpulmonary tumors, BAC occurs at younger ages, is morecommon in females (10, 11), appears to be unrelated tosmoking (4, 5), and may be increasing more rapidly (2).In accordance with the WHO (12) classification of lungcarcinoma (13), epidemiological studies have generallyconsidered BAC to be a variant of adenocarcinoma (14-18). To our knowledge, no case-control study of BACalone has been reported. We present the descriptiveepidemiology of BAC, using population-based incidencedata for 1973-1987 from the SEER program, along withrisk factors from histologically confirmed cases of BACidentified in a case-control study of lung cancer con-ducted in southern Louisiana.

Demographic Patterns

Methods. All newly diagnosed cases of primary carci-noma arising in the lung, bronchus, or trachea (Interna-tional Classification of Diseases for Oncology codes162.0-162.9) and diagnosed between 1973 and 1987

were selected from the registries participating in the SEERprogram. Histological categories were formed as follows:

BAC, other adenocarcinoma (papillary/villous, clear cell,mucinous, acinar cell, and adenocarcinoma, NOS), squa-mous cell, small cell, and total lung cancer. Annual age-adjusted rates were calculated by the direct method

using the 1970 U.S. population as the standard and areexpressed per 100,000 person-years.

Results. BAC is one of the least common pulmonarytumors; with an annual incidence of approximately 1.5cases/100,000, it accounts for less than 3% of all lungcancers. Table 1 shows annual age-adjusted incidencerates for BAC, adenocarcinoma, and squamous and smallcell carcinoma among whites and blacks, according tosex. For both races, rates are consistently higher amongmales; however, adenocarcinoma and BAC account formore of the total lung cancer incidence in females.

4 The abbreviations used are: BAC, bronchioloalveolar carcinoma; SEER,

Survival, Epidemiology and End Results; NOS, not otherwise specified;OR, odds ratio; CI, confidence interval.

on January 28, 2021. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

20

Adenocarcinoma #{149}. - - - - - - �

Year

Mite M�e Black M�. Wt�to F.m�e B$ack Fem�e

340 Epidemiology of Bronchioloalveolar Carcinoma

fig. 1. Age-adjusted incidence rates for BAC and adenocarcinoma in the United Slates, by race and sex, 1971 1987.

Table 1 Age adjusted annual incidence rates for lung cancer according

to histological group in the United States, 1973-1987, by race and sex

Incidence rate per 100,000

White Black

Male Female Male Female

BAC 1.7 1.3 1.8 1.0

Adenocarcinoma (ex- 16.0 8.5 23.7 9.2

cluding BAC(

Squamous 25.9 5.5 47.1 7.6Small cell 12.0 5.5 12.7 4.7Carcinoma NOS 20.5 6.9 28.3 7.3

Total lung cancer 80.7 29.6 120.1 32.2

Between 1973 and 1987, BAC incidence remained rela-tively constant except for a slight decline in rates amongwhite males (Fig. 1), while the incidence of all lungcancers, particularly adenocarcinoma, rose in all race/sexgroups. Black males have the highest incidence rates forall lung cancer histologies.

For BAC, blacks tend to be younger at diagnosis thanwhites (mean age of 61.4 for blacks versus 65.3 forwhites), but there is little age difference between malesand females (65.1 and 65.0, respectively). These findingsresemble lung cancer overall, in which blacks tend to beyounger than whites, but age varies little by sex. Exceptfor older age groups, incidence rates for BAC increasewith age and are consistently higher for males thanfemales (Fig. 2). BAC occurs very rarely under the age of40.

Case-Control Study

Study Design. Between 1979 and 1982, parallel hospital-based, case-control studies of lung, stomach, and pan-creatic cancers were conducted in Louisiana. Each study

§ 10

8:500.0

0

2

I

0.5

02

used the same design, methods, and questionnaire de-scribed previously (19). In summary, all newly diagnosed

cases of lung cancer (International Classification of Dis-eases-9 169.9) were ascertained from a review of admis-sion and pathology records of participating hospitals in23 southern and 3 northern Louisiana parishes )counties).Because we expected a large number of lung cancercases in the New Orleans triparish region, ascertainmentthere was limited to two private and two public hospitals(covering approximately 50% of hospital beds); outsidethe metropolitan area, all major hospitals were included.A total of 1649 incident lung cancer cases were identified,and an interview was conducted with 1253 cases (76%).

Histological classification was obtained from local hos-pital pathologists, with BAC being diagnosed in 33 (2.6%)

of the participating cases. Slides from these cases and 10randomly selected cases with other lung histologies were

later reviewed by a single pathologist using the followingdiagnostic criteria for BAC: (a) no evidence of primaryadenocarcinoma elsewhere; (h) absence of a centralbronchial carcinoma; (c) growth of malignant cells alongthe alveolar walls; and (d) intact interstitial framework ofthe lung (20). The reviewing pathologist was blinded tothe original histological diagnosis.

Controls were matched to cases by hospital, race,sex, and age within 5 years. Patients with an admittingdiagnosis of emphysema, chronic bronchitis, chronic ob-structive pulmonary disease, or cancers of the larynx,oral cavity, esophagus, or bladder were excluded from

control selection. In all, 1274 controls (85% of thoseidentified) completed an interview. For this analysis, we

chose up to five controls for each case who were com-parable with respect to race, sex, type of respondent

(self versus proxy), parish of residence (or urban versus

rural if no closer match could be found), and age within5 years.

Participants were questioned in the hospital bytrained interviewers. The questionnaire covered smok-



15 46

Age

wt�t. MaI B1�k M�s Wtite F.m�. Black Fem�s

00 75

Cancer Epidemiology, Biomarkers & Prevention 341

12

10

80

00.0

0

0

CC

I

4

2

Fig. 2. Age-specific incidence rates for BAC in the United States, by race and sex, 1973-1987.

ing, occupational and residential history, usual adult diet,alcohol use, family and personal history of disease, andexposure to materials in the workplace and during lei-sure-time activities. Subjects were asked to list all jobsheld for at least 6 months. For those too ill to respond ordeceased, interviews were conducted with the next ofkin, with priority given to a spouse or adult relative livingwith the subject. Smoking and dietary results for all lungcancers combined have been presented elsewhere (21,22).

Statistical Methods. The measure of association betweendisease and exposure was the relative risk as estimatedby the OR. Matching by age, race, sex, parish of resi-dence (or, if not feasible, urban versus rural), and type ofrespondent (self or proxy) was maintained throughoutthe analysis, using conditional logistic regression modelsto calculate both crude and adjusted ORs (23). Adjust-ment for cigarette use was based on the number of packssmoked as follows: less than 1, 1 to 1.5, and more than1 .5/day. To test for dose-response effects, P values wereobtained from the logistic model.

Descriptive Findings. Slides from a total of 43 cases(based on hospital pathology reports, 33 diagnosed withBAC and 10 with other histologies) were selected forreview, and BAC was confirmed for 21 cases. For theremaining cases, histology was as follows: 6 adenocarci-noma, acinar type; 5 squamous, 3 small cell and 2 largecell carcinoma; 1 adenocarcinoma NOS; 3 with no tumorevident on the available sections; and 2 with no slidematerial available. Because of the controversy in thediagnosis of BAC, findings are presented from the 21verified cases, all of whom were originally thought tohave BAC.

We attempted to match 5 controls to each case;however, in two instances, only 2 and 4 controls, respec-tively, fit the matching criteria. Thus a total of 101 con-trols were selected, the majority of whom were seen forcardiovascular disease (32%), followed by gastrointestinaldisorders (24%) and ill-defined symptoms or injury (13%).

Since smoking was not an a priori risk factor for BAC,subjects with smoking-related diseases were not ex-cluded from the pool of potential controls.

The race and sex distribution of study participants isshown in Table 2. Nearly 40% of the respondents werefemale, while 53% were white. Surrogate interviews wereconducted for 4 cases and 16 controls. Persons with arelatively high income had more than twice the risk ofthose with a low income (P value for trend <0.05), yetcases had fewer years of education than controls (OR =

0.6). Of all the lung cancers in the case-control study,BAC accounted for 1 .2% of the male cases versus 2.8%

of the female cases interviewed.

Smoking. Table 3 shows the distribution of cases andcontrols by smoking characteristics. Only 14% of theBAC cases (1 male, 2 females) reported never smokingcigarettes, compared to 31% of the controls. The OR forever smoking was 4.0. Risks for the amount smokedshowed a dose response, reaching 10-fold for thosesmoking more than 15 packs/day (P value for trend,<0.05). Cases tended to smoke longer than controls, withthose smoking more than 45 years having 5 times therisk of nonsmokers (trend not significant). Cessation ofcigarette smoking did not reduce cancer risk. Smokingresults were similar for males and females but are notpresented due to small numbers.

Risks varied according to the type of cigarette andwas highest for smoking nonfiltered cigarettes only. TheORs were 5.6, 1 .4, and 4.3 for smoking only nonfiltered,only filtered, and both filtered and nonfiltered cigarettes,respectively. The 2-fold risks associated with cigars andchewing tobacco were not significant, and no casesmoked a pipe or used snuff. Among the 3 nonsmokingBAC cases, none chewed tobacco or smoked a cigar, but2 lived with a smoker (a father and a spouse). For 2 ofthe 3 nonsmoking cases, information was provided by aproxy respondent.

Occupation. ORs for selected occupations, adjusted forcigarette smoking, are presented in Table 4 for men and




Table 2 Distribution of BAC cases and controls by sociodemographic

characteristics, Louisiana study, 1979-1982

Cases Controls Odds ratio

n % n � (95%CI(”

Male 13 62 65 64

White 12 57 53 53Self-respondent 17 81 83 84

Age

<55 5 24 25 25

55-64 7 33 38 38

65 or older 9 43 38 38

Meanage 63.4 62.8

Education9yearsormore 5 20 0.6(0.2, 1.9)

Income (thousand/

year(

<5 6 25 1.0 (referent(5-9 1 28 0.1 (0.01, 0.91

10-24 5 23 1.3 (0.3, 5.7(

25 or more 8 19 2.6 (0.5, 12.9)

Unknown 1 6

., Odds ratios adjusted for the number of packs of cigarettes smoked perday.

1� Column percentages do not sum to 100 due to rounding.

women combined. Since none of the nonsmoking casesheld these jobs, we were unable to evaluate occupationalassociations independently. The risk for ever-employ-ment in motor freight occupations was of borderlinesignificance (OR = 3.8) and increased to 4.4 for 10 ormore years of employment (P value for trend <0.05).Excess but nonsignificant risks were found for ever-em-ployment in construction occupations (OR = 1.4),sugarcane farming (OR = 6.8), and working in petroleummanufacturing (OR = 1 .8). ORs for construction workersrose to 2.9 for 10 or more years of employment (trendnot significant), and the two cases working in sugarcanefarming did so for at least 10 years. Slight, nonsignificantexcesses were observed for farming (OR = 1.2) andworking in oil or gas extraction (OR = 1 . 1 ). The one case

who reported being a mechanic (OR = 1 .9) also workedin construction and motor freight occupations. A slight

deficit in risk was observed for shipbuilding workers (OR= 0.8).

Medical History. Odds ratios for prior illnesses and can-cer in family members are presented in Table 5. Self-

reports of pulmonary diseases were not verified but wereuncommon in this study group. Significantly more casesthan controls had emphysema (OR = 4.1), and all werecigarette smokers. Only one case reported a history ofpneumonia, and two cases had asthma (OR = 3.9). Ahistory of cancer was more common in families of casesthan controls; the risk was 2-fold for having two or morefirst-degree relatives with any cancer and rose to 2.5 forlung cancer in at least one family member. Familial cancerwas not reported by the three nonsmoking cases.

Discussion

To our knowledge, this lung cancer case-control study isthe first to report risk factors specifically for BAC andincludes only cases the histology of which could beverified by a pathologist skilled in the diagnosis of this

Table 3 Odds ratios for BAC by smoking variables, Louisi,ina study,

1979- 1982

Exposure Cases Controls Odds ratio 195% CI(

Cigarettes

Never smokedEver smoked

318

3170

1.0 (referent)’4.0 (0.9, 18.8)

Packs/day

<11-1�/2

>1�/2

567

312810

2.7(0.5, 15.4)3.9(0.7, 22.2)

10.0(1.8, 57.1)

Years

1-30

31-40

41-4546 or more

4

4

46

18

16

1719

3.5(0.6, 20.91

3.8(0.6,23.51

4.5 (0.7, 31.015.4 (0.9, 32.11

Never stopped

Former smoker

Years quit

1

2-9lOormore

6

12

435

3733

138

12

2.5 (0.5, 13.315.8 (1.2, 29.1)

4.9(0.8, 30.7)6.0(0.8,42.5(6.7(1.1,41.01

Cigarette typeNonfilteredFiltered

Both

61

1 1

208

42

5.6 (0.9, 33.6)1.4(0.1, 18.9)

4.3 (0.9, 20.9)

Cigars 5 14 2.5 (0.7, 9.3)

Chewingtobacco 3 9 2.0(0.4, 10.1)

., Except for packs per day, odds ratios adjusted for the number of packsof cigarettes smoked per day.

tumor. Based on this review, more than 20% of the lungcancers with an original hospital-based diagnosis of BACwere subsequently reclassified as adenocarcinoma (6acinar tumors and 1 adenocarcinoma, NOS). This under-scores the difficulty in distinguishing this rare tumor andsuggests that despite the high rate of microscopic confir-mation of lung cancers in the SEER registry (including98% of the BAC cases), our descriptive analysis of BACpatterns should be interpreted cautiously. In the case-control study, the added slide review avoided diseasemisclassification, but findings must be viewed as prelim-mary because of the small number of cases.

Unlike findings from some clinical surveys of BAC(4, 5), our study showed an association with cigarette

smoking that is consistent with the moderate smokingrisks observed for adenocarcinoma in case-control stud-

ies (15-18, 22, 25-27). Nearly all the cases in our studysmoked cigarettes at some time in their adult life, and itis notable that interviews were conducted by proxy for

two of the three BAC cases who reportedly neversmoked. ORs rose as the amount and duration of smok-ing increased but did not attain the level of risk associatedwith squamous and small cell lung cancers (15-18, 22,25-27). Smokers of nonfiltered cigarettes were at greatest

risk, followed by those smoking both filtered and nonfil-tered cigarettes. The risks for smoking variables may beunderestimated, however, in light of our use of hospital-ized controls, including those diagnosed with smoking-related conditions. Smoking cessation did not lower risk,since nearly one-half of the former smoking cases hadquit for at least 10 years. The role of passive smokingcould not be evaluated due to the small sample size and



Cancer Epidemiology, Biomarkers & Prevention 343

Table 4 Odds ratios for BAC by selected occupations, Louisiana study,1979-1982

Occupation Cases Controls Odds ratio )95% Ct)”

Motor freightEver

<10 years

Employed 10+ years

6

3

3

9

5

4

3.8 (1.0, 14.6)

3.5 (0.7, 18.0)

4.4 (0.7, 29.3)

Construction

Ever

<l0yearsEmployed 10+ years

5

23

1 5

96

1 .4 (0.4, 5.4)

0,8 (0.1, 4.8)2.9 (0.5, 17.3)

Mechanic, repairman 1 4 1.4 (0.1, 17.3)

FarmingSugarcane

42

231

1.2 (0.3, 4.8)6.8 (0.5, 85.2)

Oil or gas extraction 2 8 1 . 1 (0.2, 6.1)

Petroleum manufacturing 2 7 1.8 (0.3, 10.0)

Shipbuilding 2 1 1 0.8 (0.2, 4.4)

‘, Odds ratios adjusted for th e numbe r of packs of cigarettes smoked perday.

the small number of nonsmoking cases. Few respondentsused other forms of tobacco, including pipes, cigars, orsmokeless tobacco.

The risk of lung cancer has been related to severaloccupational exposures, including asbestos (17, 28-30),polycyclic hydrocarbons (31, 32), and recently to wooddusts (19, 33, 34) and motor exhausts (35, 36). Whilethere are few clues to occupational determinants of BAC,an excess risk of lung adenocarcinoma has been reported

for carpenters, motor vehicle drivers, and mechanics andrepairmen (37). We found significantly high risks for long-term employment in motor freight and construction jobs,where contact with asbestos may occur; however, allexposed cases also smoked cigarettes. The risk for me-

chanics and repairmen was slightly elevated but wasbased on a single case who also worked in constructionand motor freight. Recently, bronchioloalveolar neo-plasms were induced in mice following inhalation oftetranitromethane, an oxidizer used as a diesel fuel ad-ditive that enhances octane ratings (38). Consistent with

a lung cancer association found in an earlier death certif-icate survey in Louisiana (39), we found a nonsignificantexcess risk for sugarcane farmers. Unlike other studies oflung cancer in coastal areas (29), we found no excess risk

associated with the shipbuilding industry, where asbestosexposure may be heavy. An excess of lung adenocarci-noma has been found among plumbers, printers, andpainters (37), but no cases in our study held these jobs.

An etiological role for fibrosing lung disorders in BAChas been suspected in several clinical series (2-4). Epi-

demiological studies have linked lung adenocarcinomato antecedent pneumonia and tuberculosis (14, 25),while emphysema has been associated with squamous

and oat cell carcinoma (25). In our study, where reportsof prior lung diseases were uncommon and not verified,a significant association was observed for emphysemabased on three cases, all smokers. However, this risk mayhave been inflated due to control group restrictions, inparticular, the exclusion of those hospitalized withchronic pulmonary disease. Familial occurrences of BAChave been reported (40, 41), and in our study the parents

Table 5 Odds ratios for BAC by history of respiratory disease and

cancer in family members, Louisiana study, 1979-1982

Cases Controls Odds ratio (95% Cl)”

Asthma 2 3 3.9 (0.6, 24.5)Emphysema 3 3 4.1 (0.7, 22.1)Pneumonia 1 3 1.7 (0.2, 20.5)

Family cancerAny form

1 member 7 36 1.0 (0.3, 3.0)2 or more members 2 6 2.0 (0.4, 1 1.9)

Lungcancer 3 6 2.5 (0.5, 11.7)Father 1 2 3.9 (0.2, 72.2)

Sibling 2 4 2.1 (0.4, 13.2)

., Odds ratios adjusted for the number of cigarette packs smoked per day.

and siblings of cases were more likely to have had lungcancer than those of controls. No attempt was made toverify the diagnosis of cancer in a family member.

BAC is morphologically similar to sheep pulmonaryadenomatosis (jaagsiekte), which appears to be causedby infection with a retrovirus (42, 43). Animal-to-humantransmission has not been established, although caseexposure to diseased sheep has been reported (2). Be-cause sheep rearing is not common in Louisiana, ourinterview did not include relevant questions; however,farmers were not at increased risk. Tumors with a viraletiology tend to be diagnosed at younger ages, yet wefound that the age distribution of BAC cases resembledthat of all lung cancers reported to the SEER program.

Incidence data showed higher rates of BAC andadenocarcinoma among men, but both tumors wererelatively more common in women; likewise, in our case-control study, BAC accounted for more lung cancer casesamong women than men. Smoking patterns may explainpart of the gender differences in the distribution of lungcancer histologies (15), although men and women sharedsimilar risks for smoking variables in our study. A role forendogenous hormonal factors and cooking oil vapors hasbeen suggested in some studies of lung adenocarcinoma(14, 25), but our interview did not include relevantquestions.

In summary, many of the epidemiological featuresof BAC presented in this report resemble those of ade-nocarcinoma, supporting the classification of BAC as avariant of adenocarcinoma. Like adenocarcinoma, BACis relatively more frequent among women, and the smok-ing risks are lower than those for squamous or small cellcarcinoma. Although most cases in our study could beattributed to cigarette smoking, this finding should beviewed cautiously due to the small sample size. Risks forsome occupational groups, notably truck drivers and

construction workers, were elevated, but smoking mayhave confounded the associations. Similarly, the excessrisks associated with emphysema and familial cancer

could not be evaluated apart from smoking. Neverthe-less, these findings provide clues for larger case-controlstudies aimed at clarifying the environmental and hostdeterminants of BAC.

References

1. National Cancer Institute. Cancer Statistics Review 1973-1981, DHHSPublication No. (NIH) 90-2789. Washington, DC: U.S. Government Print-ing Office, 1990.




22. Fontham, E. T. H., Pickle, L. W., Haenszel, W., Correa, P., Liii, Y.,

2. Edwards, C. W. Alveolar carcinoma: a review. Thorax, 39: 166-174,

1984.

3. Shraufnaget, D., Peloquin, A., Pare, I. A. P., and Wang, N. S. Differ-entiating bronchioloalveolar carcinoma from adenocarcinoma. Am. Rev.

Respir. Dis., 125: 74-79, 1982.

4. Greco, R. I,, Steiner, R. M., Goldman, S., Cotler, H., Patchefsky, A.,

and Cohn, H. E. Bronchoalveolar cell carcinoma of the lung. Ann. Thorac.

Surg., 41: 652-656, 1986.

5. Watson, W. L., and Farpour, A. Terminal bronchiolar or “alveolar cell”cancer of the lung. Two hundred sixty-five cases. Cancer (Phita.), 19:

776-780, 1960.

6. Bennett, D. E., and Sasset, W. F. Bronchiolar carcinoma: a valid

clinicopathologic entity? A study of 30 cases. Cancer (Phila,(, 24: 876-

887, 1969.

7. Donaldson, I, C., Kaminsky, D. B., and Elliott, R. C. Bronchiolarcarcinoma. Report of 1 1 cases and review of the literature. Cancer (Phila.(,

4!: 250-258, 1978.

8. Thomas, I. St. I.. Tullett, W. M., and Stack, B. H. R. Bronchioloalveolarcell carcinoma: a 21-year retrospective study of cases at the Western

Infirmary, Glasgow. Br. I. Dis. Chest, 79: 132-140, 1985.

9. Kennedy, A. Relationship between cigarette smoking and histologicaltype of lung cancer in women. Thorax, 28: 204-208, 1 973.

10. Liebow, A. Bronchiolo-alveolar carcinoma. Adv. Intern. Med., 10:329-58, 1960.

11. Storey, C. F., Knudtson, K. P., and Lawrence, B. I. Bronchiolar

(“alveolar cell”) carcinoma of the lung. I. Thorac. Surg., 26: 331-403,1953.

12. World Health Organization. International Classification of Diseases.

Manual of the International Statistical Classification of Disease, Inturiesand Causes of Death, Rev. 9, Vols. 1 and 3. Geneva: World HealthOrganization, 1978.

13. Kreyberg, L. Histological Typing of Lung Tumours, Vol. 1. Inlerna-

tional Histological Classification of Tumours. Geneva: World Health Or-ganization, 1967.

14. Wu, A. H., Henderson, B. E., Pike, M. C., and Yu, M. C. Smoking

and other risk factors for lung cancer in women. I. Natl. Cancer Inst., 74:

747-751, 1985.

1 5. Lubin, I. H., and Blot, W. I. Assessment of lung cancer risk factors by

histologic category. I. NatI. Cancer Inst., 73: 383-389, 1984.

16. Damber, L. A., and Larsson, L. G. Smoking and lung cancer with

special regard to type of smoking and type of cancer. A case-controlstudy in north Sweden. Br. I. Cancer 53: 673-681, 1986.

17. Vena, I. E., Byers, T. E., Cookfair, D., and Swanson, M. Occupation

and lung cancer risk. An analysis by histologic subtypes. Cancer (Phila.),56: 910-917, 1985.

18. Lam, T. H., Kung, I. T. M., Wong, C. M., Lam, W. K., Kleevens, I. W.

L., Saw, D., Hsu, C., Seneviratne, S., Lam, S. Y., Lo, K. K., and Chan, W.

C. Smoking, passive smoking and histological types in lung cancer inHong Kong Chinese women. Br. I. Cancer, 56: 673-678, 1987.

19. Correa, P., Pickle, L. W., Fontham, E. T. H., Dalager, N. A., andHaenszel, W. The causes of lung cancer in Louisiana. In: M. Mizell andP. Correa (eds.), Lung Cancer: Causes and Prevention, pp. 73-82. Deer-

field Beach, FL: Verlag Chemie International, Inc., 1984.

20. Greenberg, S. D., Smith, N. M., and Spjut, H. I. Bronchioloalveolarcarcinoma-cetloforigin. Am. I. Clin. Pathol., 63: 153-167, 1975.

21. Correa, P., Pickle, L. W., Fontham, E. T. H., Lin, Y., and Haenszel,

‘vV. Passive smoking and lung cancer. Lancet, 2: 595-597, 1983.

and Falk, R. T. Dietary vitamins A and C md lung can er risk in Louisiana.

Cancer (Phila.), 62: 2267-2273, 1988.

23. Preston, D. L., Lubin, I. H., ,ind Pierce, D. A. EPICURE. Se,sttte, WA:

HiroSoft International Corp., 1 99 1

24. Knudson, R. I., Hatch, H. B., Mit hell, M. T., Ir., ,ind O hsner, A.

Unusual cancer 01 the lung. II. Bron hiolar carcinom,i 01 the lung. Dis.

Chest, 48: 628-631, 1965.

25. Gao, Y. T., Blot, W. I., Zheng, SN., Ershow, A. G., Hsu, C. W., Levin,L. I., Zhang, R., ,ind Fraumeni, I. F., Ir. lung c,ln( er ,lnlong Chinese

women. tnt. I. Cancer, 40: 604-609, 1987.

26. Gao, Y. T., Blot, W. I.. Zheng, ‘A’., and Fraumeni, I. F., (r. Lung cancerand smoking in Shanghai. tnt. F. Epidemiol.. 1 7’: 277 � 280, 1988.

27. Stayner, L. T., and Wegnian, U. H. Snioking, occupation, and histo-

pathology of lung c,incer: a case-c ontrol study with the oct of the ThirdNational Cancer Survey. I. NatI. Cancer Inst., 70: 421 -426, 1983.

28. Hinds, M. W., Kolonel, L. N., and tee, I. Appli ation of a job-

exposure matrix to a c,sse-control study 01 lung cancer. I. NaIl. CancerInst., 75: 193-197, 1985.

29. Blot, W. I.. Harrington, I. M., Toledo, A., Hoover, R. N., Heath, C.

W., and Fraumetii, I. F., (r. Lung caticer alter employment in shipyards

during World War II. N. EngI. I. Med., 299: 620-624, 1978.

30. Blot, w. I., Morris, L. E., StrouI�’, R., Tagnon, I., and Fr,iumeni, I. F.,

Jr. Lung and laryngeal c,incers in relatiiin 10 shipyard c’niploynient in

coastal Virginia. I. NaIl. Cancer InsI., 65: 571 -575, 1980.

3 1 . Lloyd, I. W. Long-term mortality study (it steelworkers. V. Respiratory

cancer in coke plant workers. I. 0c lip. Med., 13: 51-68, 1971.

32. Blot, W. I., Brown, L. M., Pottern, F. M., Stone, B. I.. and Fraumeni,

I. F., Ir. Lung cancer among long-term steel workers. Ani. I. Epidemiol..1 17: 706-716, 1983.

33. Levin, L. I., Zheng, W., Blot, W. (., Gao, Y. T., and Fraiimeni, ). F., (r.

Occupation and lung cancer in Shanghai: a case-control study. Br. (. nd.Med., 45: 450-458, 1988.

34. Blot, W. I., Davies, . E., Morris Brown, L., Nordw.itt, C. W., Buiatti,

E., Ng, A., and Fraumeni, I. F., (r. Occupation and the high risk 01 lungcancer in northeast Florida. Cancer Phita.), 50: 364-37 1 , 1982.

35. Vineis, P., Thomas, T., Hayes, R. B., Blot, W. I., M,isiiri, T. I., Pickle,L. W., Correa, P., Fonthani, E. T. H., and Scohenberg, I. l’roportion of

lung cancers in males due to occupation in ditlerent areas of the U.S. tnt.

I. Cancer, 42: 851 -856, 1988.

36. Benhamou, S., Benhamou, E., and Flamont, R. Ocupational risk

factors of lung cancer in a French case-control study. Br. I. Ind. Med., 45:231-233, 1988.

37. Zahm, S. H., Brownson, R. C., Chang, I. C., and Davis, I. R. Study oflung cancer histologic types, occup,ilion, and smoking in Missouri. Am.

I. Ind. Med., 14: 565-578, 1989.

38. Bucher, I. R., Hull, I. F., lokinen, M. P., Haseman, I. K., Stedham, NI.,and Cholakis, I. M. Inhalation of tetranitromethane causes tiasal passage

irritation and pulmonary carcinogenesis in rodents. Cancer lett., 57: 95-101, 1991.

39. Rothschild, H., and Mulvey, I. I. An increased risk for lung cancermortality asociated with sugar,ine f,irming. I. NatI. Cancer Inst., 68: 755-

760, 1982.

40. loishy, S. K., Cooper, R. A., and Rowley, P. T. Alveolar cell clriiiini,iin identical twins. Ann. Intern. Med., 87: 447-450, 1977.

41 . Paul, S. M., Bacharach, B., and Goepp, C. A genetic influen e on

alveolar cell carcinoma. I. Surg. Oncol., 16: 249-252, 1987.

42. Verwoerd, D. W. Biotechnology, vir,il oncogenesis and jaagsiekte.Onderstepoort I. Vet. Res., 52: 14.5- 148, 198.5.

43. Perk, K., and Hod, I. Sheep lung carcinoma: an endemic analogue of

.1 sporadic human neoplasm. I. NatI. C,incer Inst., (,9: 747. 749, 1982.



1992;1:339-344. Cancer Epidemiol Biomarkers Prev R T Falk, L W Pickle, E T Fontham, et al. Epidemiology of bronchioloalveolar carcinoma.

Updated version

http://cebp.aacrjournals.org/content/1/5/339

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/1/5/339To request permission to re-use all or part of this article, use this link



http://cebp.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Epidemiology of Bronchioloalveolar Carcinoma1characteristics of lung adenocarcinoma. Received 10/24/91. ‘ Supported by Contract N01-CP-91023 from the National Cancer Institute. To