Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 28.
Published in final edited form as: J Natl Cancer Inst. 1997 Sep 17;89(18):1366–1373. doi: 10.1093/jnci/89.18.1366

Molecular Damage in the Bronchial Epithelium of Current and Former Smokers

Ignacio I Wistuba 1, Stephen Lam 1, Carmen Behrens 1, Arvind K Virmani 1, Kwun M Fong 1, Jean LeRiche 1, Jonathan M Samet 1, Sudhir Srivastava 1, John D Minna 1, Adi F Gazdar 1,*
PMCID: PMC5193483  NIHMSID: NIHMS834695  PMID: 9308707

Abstract

Background

Most lung cancers are attributed to smoking. These cancers have been associated with multiple genetic alterations and with the presence of preneoplastic bronchial lesions. In view of such associations, we evaluated the status of specific chromosomal loci in histologically normal and abnormal bronchial biopsy specimens from current and former smokers and specimens from nonsmokers.

Methods

Multiple biopsy specimens were obtained from 18 current smokers, 24 former smokers, and 21 nonsmokers. Polymerase chain reaction-based assays involving 15 polymorphic microsatellite DNA markers were used to examine eight chromosomal regions for genetic changes (loss of heterozygosity [LOH] and microsatellite alterations).

Results

LOH and microsatellite alterations were observed in biopsy specimens from both current and former smokers, but no statistically significant differences were observed between the two groups. Among individuals with a history of smoking, 86% demonstrated LOH in one or more biopsy specimens, and 24% showed LOH in all biopsy specimens. About half of the histologically normal specimens from smokers showed LOH, but the frequency of LOH and the severity of histologic change did not correspond until the carcinoma in situ stage. A subset of biopsy specimens from smokers that exhibited either normal or preneoplastic histology showed LOH at multiple chromosomal sites, a phenomenon frequently observed in carcinoma in situ and invasive cancer. LOH on chromosomes 3p and 9p was more frequent than LOH on chromosomes 5q, 17p (17p13; TP53 gene), and 13q (13q14; retinoblastoma gene). Microsatellite alterations were detected in 64% of the smokers. No genetic alterations were detected in nonsmokers.

Conclusions

Genetic changes similar to those found in lung cancers can be detected in the nonmalignant bronchial epithelium of current and former smokers and may persist for many years after smoking cessation.

Lung cancer is the most frequent cause of cancer deaths in both men and women in the United States (1). Tobacco smoking is accepted as a major cause of cancers of the lung and of several other cancer types (2). As with other epithelial cancers, lung cancer is believed to arise after a series of progressive pathologic changes (preneoplastic lesions) in the bronchial epithelium (3). The sequential preneoplastic changes have been defined for centrally arising squamous cell carcinomas but are poorly documented for small-cell carcinomas and adenocarcinomas. Advanced preneoplastic changes occur far more frequently in smokers than in non-smokers, and these changes increase in frequency with the amount of smoking, after adjustment for age (4,5). Morphologic recovery occurs after smoking cessation (4,6), although elevated lung cancer risk persists (7). Changes in bronchial epithelium, including metaplasia and dysplasia, have been utilized as surrogate end points for chemoprevention studies (8,9). Risk factors that identify normal or premalignant bronchial tissue at risk for malignant progression need to be better defined.

Many mutations, especially those involving recessive oncogenes, have been described in lung cancers (10,11). Loss of heterozygosity (LOH) analyses utilizing polymorphic microsatellite DNA markers are frequently used to identify allelic losses at specific chromosomal loci. Allelic losses at chromosomal regions 3p, 9p, and 17p occur relatively early during the multistage development of invasive lung cancer (12–17). In addition, there is evidence of more generalized genomic instability in lung cancer and its preneoplastic lesions. Widespread aneuploidy occurs throughout the respiratory epithelium of lung cancer patients (18). Microsatellite alterations are found in many human cancers, including lung cancer, and may serve as clonal markers for early cancer detection (19,20). Microsatellite alterations involve changes in the size of the simple nucleotide repeats of polymorphic microsatellite DNA sequences, resulting in altered electrophoretic mobility of one or both alleles. In lung cancers, such alterations have been reported to occur at frequencies ranging from 0% to 45% (19,21–23). Although the mechanisms underlying microsatellite alterations are currently unknown, they may represent a form of genomic instability (24).

Most previous molecular studies of lung tissue have been performed in material from small numbers of subjects with concurrent lung cancer, and only scant information is available about molecular changes in the respiratory epithelium of smokers without cancer (15,16,25). In this study, we determined the frequency of such alterations at eight chromosomal regions, which are frequently deleted in lung cancers, in the bronchial epithelium of current and former smokers in comparison with nonsmokers.

Materials and Methods

Study Populations

We studied bronchial biopsy specimens obtained by fluorescence bronchoscopy from 63 subjects (21 nonsmokers, 18 current smokers, and 24 former smokers) (Table 1). All subjects were recruited at the British Columbia Cancer Agency, Vancouver, Canada, as part of an Institutional Review Board-approved clinical trial to study the effect of smoking on the respiratory epithelium. All participants gave written informed consent before enrollment into the study. Because of our inability to enroll older non-smokers, the control subjects consisted of 16 relatively young, healthy volunteers, three subjects who had bronchoscopy for chronic cough, and two individuals with a history of occupational exposure to asbestos. Subjects were categorized as to smoking status as follows: (a) Nonsmokers were subjects who had smoked fewer than 365 cigarettes in their lifetime, (b) current smokers were subjects who had smoked more than 365 cigarettes in their lifetime and who either were currently smoking or had stopped smoking within the previous 12 months, (c) former smokers were subjects who had smoked more than 365 cigarettes in their lifetime and who had stopped smoking for longer than the previous 12 months, and (d) lifetime smokers (also referred to as “smokers”) consisted of a combination of current and former smokers. Because almost all adults have at least a minimal exposure to cigarette smoking, we arbitrarily chose a 365-cigarette exposure, equivalent to one cigarette per day for a year, as the distinction between smokers and nonsmokers. All smokers had smoked more than 20 pack-years (pack-years = number of packs per day multiplied by the number of years of smoking), except for three subjects (1, 10, and 12 pack-years). Most former smokers (72%) had ceased smoking for 5 years or longer (mean, 11 years; range, 1–48 years). Sixteen smokers had a history of prior carcinoma in situ (CIS) or invasive lung cancer.

Table 1.

Characteristics of patients and control subjects: smoking histories, clinical information, and biopsy specimens

Subject categories No. Pack-years,*,
mean (range)		 Years of smoking
cessation,
mean (range)		 Sex,
M:F† 		Age, y
 Biopsy specimens studied
Mean (range) Median Total Mean (range)
Lifetime nonsmokers 21 NA‡ NA‡ 15:6 34 (21–60) 29 67 3.1 (1–7)
Lifetime smokers 42§ 49 (1–132) — 34:8 63 (26–88) 61 188 4.5 (2–9)
  Current smokers 18 44 (12–77) 0 14:4 60 (34–83) 59 92 5.1 (2–9)
  Former smokers 24 54 (1–132) 11 (1–48) 20:4 65 (26–88) 63 96 4.0 (2–9)
All subjects 63 — — 49:14 53 (26–88) 58 255 4.1 (1–9)
Open in a new tab
*

Pack-years = number of packs of cigarettes smoked per day multiplied by the number of years of smoking.

†

M = male; F = female.

‡

Not applicable.

§

Two current smokers and 14 smokers had a previous lung cancer.

Of the subjects, 49 (78%) were males and 14 (22%) were females. The mean age of the smokers was 63 years (range, 26–88 years). Because we were unable to recruit older nonsmokers, the average age of the nonsmokers was considerably lower (mean, 34 years; range, 21–60 years). Other relevant subject information is presented in Table 1.

Fluorescence Bronchoscopy

Fluorescence bronchoscopy was performed by one of us (S. Lam), as previously described (26,27). The procedure was performed on the subjects under local anesthesia with the use of the LIFE-Lung device following conventional white-light examination. Briefly, upon illumination of the bronchial surface with a blue light (405–442 nm), the autofluorescence intensity progressively decreases as the tissue changes from normal to metaplasia, dysplasia, CIS, and invasive cancer. Lesions as small as 1 mm in surface diameter were localized, and biopsy specimens were taken under direct vision. Specimens were taken from all areas of the bronchial tree that were suspicious for moderate dysplasia or worse. In addition, at least two biopsy specimens from widely separated areas of normal or minimally abnormal fluorescence were taken.

Identification of Preinvasive Lesions

Sections (5 µm thick) of bronchial biopsy tissues were stained with hematoxylin–eosin and examined by two reference pathologists (A. F. Gazdar and J. LeRiche) and scored with the use of published criteria for the histologic identification of epithelial premalignant lesions (3). In case of disagreement, a consensus diagnosis was achieved with help of the third pathologist (I. I. Wistuba). Pathologic diagnoses were categorized as follows: category 1, normal bronchial epithelium; category 2, hyperplasia (goblet cell or basal cell type) or simple squamous metaplasia without dysplasia; category 3, mild dysplasia; category 4, moderate or severe dysplasia; and category 5, CIS.

Microdissection and DNA Extraction

Microdissection and DNA extraction were performed from specimens mounted on microscope slides, as previously described (13). Precisely identified areas of normal and abnormal bronchial epithelia were microdissected under microscopic visualization. Microdissected stromal cells from the same slides provided constitutional DNA. Biopsy specimens containing a total of at least 300–800 epithelial cells in one or more sections were regarded as adequate for analysis and were microdissected. After DNA extraction, 5 µL of the digested samples, containing DNA from at least 50 cells, was used for each polymerase chain reaction (PCR) reaction.

Polymorphic DNA Markers and PCR-LOH Analysis

To evaluate LOH and microsatellite alterations, we used primers flanking dinucleotide and multi-nucleotide microsatellite repeat polymorphisms located at the following genes or chromosomal locations: 3p14.2 (FHIT [fragile histidine triad] locus, D3S4103), 3p14.3–21.1 (D3S1766), 3p21 (D3S1447, D3S1478, and D3S1029), 3p22–24.2 (D3S2432, D3S1351, and D3S1537), 5q22 (L5.71CA), 9p21 (D9S171 and IFNA [interferon alfa]), 13q14 (RB [retinoblastoma] gene, dinucleotide repeat at intron 2, and tetranucleotide repeat at intron 20), and 17q31.1 (TP53 [p53] gene, the dinucleotide repeat TP53, and a pentanucleotide repeat). With four exceptions (28–30), most of the primer sequences were obtained from the Human Genome Database. Allelic loss was determined by modifying a previously described (13) PCR-based assay, generating 32PO4-labeled amplification products, as follows: (a) Nested PCR methods were used; (b) for detection of loci within the TP53 and FHIT genes and the 3p21 region, hot-start PCR (TaqStart Antibody; CLONTECH Laboratories, Inc., Palo Alto, CA) was used. In individual subjects, only informative markers that demonstrated constitutional heterozygosity were tested for LOH. There were no statistically significant differences in heterozygosity rates according to chromosomal regions for any subject category. LOH was scored by visual detection of complete absence of one allele. Micro-satellite alterations were detected by a shift in the mobility of one or both alleles (Fig. 1). After generation of the initial data and breaking of the histologic code (see below for information on binding), we confirmed that LOH and microsatellite alterations were not artifacts by re-examining all changes in histologically normal biopsy specimens by repeat PCR analysis, frequently from an independent microdissection.

Fig. 1.

Fig. 1

Open in a new tab

Representative autoradiographs of microsatellite DNA analyses involving biopsy specimens from four smoker subjects (a–d), showing loss of heterozygosity at chromosomal regions 3p22-24.2 (a), 3p14.2 (FHIT gene) (b), 17p (TP53 gene) (c), and 9p21 (d). S = normal stromal cells; 1 = histologic category 1 (normal bronchial epithelium); 2 = histologic category 2 (hyperplasia or simple squamous metaplasia without dysplasia); 3 = histologic category 3 (mild dysplasia); 4 = histologic category 4 (moderate or severe dysplasia); 5 = histologic category 5 (carcinoma in situ). In panel c, in the TP53 pentanucleotide repeat marker, a microsatellite alteration is evident in normal epithelium (category 1, lane 2). See text for more details.

Calculation of Indices for Documenting Extent of Molecular Changes

Because heterozygosity at the different loci varied between subjects, the number of chromosomal regions tested in subjects varied. Thus, indices were created to compare molecular changes between subjects and between biopsy specimens. The fractional regional loss index for individual biopsy specimens (FRL-biopsy) and the fractional regional loss index for all biopsy specimens from an individual subject (FRL-subject) were calculated as follows:

FRL-biopsy index=total number of chromosomal regions with LOHtotal number of informative,for any one biopsy.
FRL-subject index=total number of chromosomal regions with LOH summed for alltotal number of informative summed for all,for any one subject.

Statistical Analyses

Pathologists and laboratory staff were blinded as to subject category and other clinical information until the data were merged for analysis. Data were analyzed by use of chi-squared methods for proportions (31). Because of the distribution of the biopsy and patient indices, the nonparametric Wilcoxon test (31) was used to compare the groups. Spearman correlation coefficients were calculated between the number of regions with LOH and age and pack-years of smoking (31). The cumulative binomial test (32) was used to examine the likelihood that a particular event (loss of the same allele in paired biopsy specimens) occurs at a particular probability when observed in repeated trials. When the results are compared with a chance occurrence or nonoccurrence, the particular probability of comparison is .5. All reported P values are two-sided.

Results

Histologic Changes

A total of 315 biopsy specimens were obtained from 63 subjects (average, five per subject; range, one to nine per subject). Two hundred fifty-five specimens (81%) contained adequate numbers of surface epithelial cells to perform multiple DNA analyses (Table 1). Ninety-two biopsy specimens were from current smokers (mean of 5.1 per subject), 96 were from former smokers (mean of 4.0), and 67 were from nonsmokers (mean of 3.1). While the nonsmokers were significantly younger than the smokers, sample adequacy was similar in smokers and nonsmokers (Table 1).

The distribution of histologic categories of biopsy specimens was significantly different (P<.001) between smokers and nonsmokers (Fig. 2, A). In nonsmokers, 65 (97%) of 67 biopsy specimens demonstrated normal or slightly abnormal changes (categories 1 and 2), two (3%) biopsy specimens demonstrated mild dysplasia (category 3), and none demonstrated more severe changes (categories 4 or 5). In contrast, 96 (51%) of 188 biopsy specimens from 33 (79%) of 42 smoker subjects demonstrated histologic categories 3–5, and only 26 (14%) of 188 biopsy specimens were normal. In addition, only four (4%) of 92 biopsy specimens from current smokers were normal compared with 24 (25%) of 96 biopsy specimens from former smokers (P<.001).

Fig. 2.

Fig. 2

Open in a new tab

A) Loss of heterozygosity (LOH) in individual biopsy specimens according to smoking status and histologic categories. LOH is expressed in terms of the fractional regional loss biopsy (FRL-biopsy) index (i.e., the fraction of chromosome regions showing LOH in each biopsy specimen) (range, 0–1). Horizontal bars indicate the mean for each histologic category. B) The FRL-subject (i.e., fractional regional loss for all biopsy specimens from an individual subject) index distribution in current and former smoker subjects. Horizontal bars represent the mean for each group of subjects. See text for more details.

Comparison of Molecular Changes Between Nonsmokers and Smokers

A most striking finding was the complete absence of molecular changes in every biopsy specimen from every non-smoker subject. In contrast, a very high frequency of LOH at one or more chromosomal regions was detected in 36 (86%) of 42 smokers and in 91 (48%) of 188 of their biopsy specimens (P = .0001) (Figs. 2 and 3). There was a modest correlation between the number of molecular changes per subject (FRL-subject index) and smoking exposure (r = .34). Of interest, the difference in the mean FRL-subject index between current (0.19) and former smokers (0.18) was not significant (P = .98) (Fig. 2, B). In addition, the difference in the mean FRL-subject index between subjects with a history of prior lung cancer (0.20) and those without such a history (0.18) was not significantly different (P = .12).

Fig. 3.

Fig. 3

Open in a new tab

A) Relationship between loss of heterozygosity (LOH) at individual chromosomal regions according to histologic categories in smokers. B) LOH at specific regions of chromosome 3p according to histologic categories in smokers. See legend to Fig. 1 for histologic category definitions.

Of interest, 10 subjects (five current smokers and five former smokers; 24% of the subjects who smoked) demonstrated LOH at one or more chromosomal regions in all of their biopsy specimens. In contrast, in six subjects (one current smoker and five former smokers), no LOH was detected in any biopsy specimen (Fig. 2, B).

Because the smokers tended to be older and, in fact, there were no non-smokers above 60 years of age, we attempted to use various multivariate models to control for the effect of age so that the effect of smoking could be estimated without confounding by the age differences. Because of the differences in the age distributions, these models were not fully successful for this purpose. However, none of the three nonsmokers over 45 years of age had any mutations. In contrast, among four smokers under 45 years of age, all had multiple mutations in multiple biopsy specimens.

Correlation Between Molecular Alterations and Histologic Changes in Smokers

We correlated the fraction of regional loss per biopsy specimen with the histologic category in smokers. The data, summarized in Fig. 2, A, demonstrate that the mean FRL-biopsy indices were similar (0.13–0.15) from category 1 (normal epithelium) to category 4 (moderate or severe dysplasia) until a significant rise (0.61) occurred in category 5 (CIS) (P = .001).

Another important observation was the presence of frequent LOH in histologically normal biopsy specimens (category 1; Fig. 2, A). LOH at one or more chromosomal regions was detected in 13 (50%) of 26 histologically normal biopsy specimens taken from 10 (53%) of 19 smokers.

LOH was detected more frequently at certain chromosomal sites than at others (Fig. 3, A and B). The most frequent allelic losses occurred at one or more chromosome 3p regions (38% of all biopsy specimens) and at chromosome 9p21 (23% of all biopsy specimens). The least frequent change was LOH at chromosome 5q (2% of all biopsy specimens, none of which were in normal epithelium). LOH at the TP53 (12% of biopsy specimens) and RB (18%) genes occurred at intermediate frequencies.

We tested whether there were differences between the four specific chromosome 3p regions (3p14.2 [FHIT gene], 3p14.3, 3p21, and 3p22–24.2) suspected to harbor tumor suppressor genes. Apart from 3p14.2 (FHIT gene), LOH at the other chromosome 3p regions was detected in histologically normal epithelium and was quantitatively similar until the CIS stage, when a statistically significant increase in frequency was noted for all regions (Fig. 3, B). In contrast, LOH at 3p14.2 (FHIT gene) was first detected at the later stage of mild dysplasia.

Allele-Specific Loss

We have previously described the phenomenon that we labeled allele-specific loss (ASL) indicating that the identical allele is lost in widely separated areas of the respiratory epithelium (13,14). In this study, we found the same phenomenon when two or more biopsy specimens from the same subject demonstrated losses of the same marker(s). In such comparisons of paired biopsy specimens, ASL was noted in 118 (93%) of 127 cases. According to the cumulative binomial test, the probability of this finding happening by chance is 1 × 10−25.

Microsatellite Alterations in Smokers

A high frequency of microsatellite alterations at one or more chromosomal loci was detected only in smokers (27 [64%] of 42 subjects and 46 [24%] of 188 biopsy specimens). Of interest, microsatellite alterations at one or more chromosomal loci were detected in four (15%) of 26 histologically normal biopsy specimens occurring in four (21%) of 19 smokers. The frequencies of microsatellite alterations did not alter with increasing histopathologic changes. There were no statistically significant differences in the frequencies of microsatellite alterations between current and former smokers or between subjects with or without a history of prior carcinoma.

Discussion

Because most lung cancers are attributable to smoking, we investigated molecular changes in the normal and abnormal bronchial epithelia of smokers and nonsmokers. Fluorescence bronchoscopy was used to identify and to obtain biopsy specimens from multiple areas of histologically normal and abnormal bronchial epithelia in nonsmokers, current smokers, and former smokers. After careful micro-dissection of the epithelium, extracted DNA was analyzed by PCR for LOH at eight chromosomal regions frequently deleted in lung cancers. These analyses utilized 15 polymorphic microsatellite markers. In nonsmokers, almost all (97%) of the biopsy specimens demonstrated normal or slightly abnormal histology. In contrast, biopsy specimens from both current and former smokers demonstrated the entire spectrum of preneoplastic pathologic changes associated with lung cancer. A significantly higher percentage of biopsy specimens from former smokers (25%) than from current smokers (4%) had normal histology. These morphologic findings are consistent with previous observations that dysplasia and CIS occur less frequently in nonsmokers than in smokers (4) and that histologic recovery of the bronchial epithelium may occur relatively rapidly after smoking cessation (4,6).

While the 21 nonsmokers consisted of healthy volunteers as well as subjects being investigated because of chronic cough or occupational exposure to asbestos, no molecular changes (either LOH or micro-satellite alterations) were present in any of the 67 biopsy specimens analyzed. In contrast, extensive LOH, frequently of multiple regions, was present in nearly half (48%) of the biopsy specimens from smokers, including 50% of histologically normal biopsy specimens. Most smokers (86%) had allelic loss in at least one biopsy specimen, and 10 subjects (24%) demonstrated allelic loss in every biopsy specimen analyzed. In addition, microsatellite alterations were present in at least one biopsy specimen from 64% of the smokers.

Interpretation of the findings on smoking is potentially limited by the differing age distribution of the smokers and non-smokers included in the study. The smokers tended to be older; in fact, there were no nonsmokers above 60 years of age. We attempted to use various multivariate models to control for the effect of age so that the effect of smoking could be estimated without confounding by the age differences. Because of the differences in the age distributions, these models were not fully successful for this purpose. However, the complete absence of genetic changes in the nonsmokers across the age span from 20 to 60 years indicates that the substantial genetic changes found in the smokers cannot be attributed to age alone. Even in the older nonsmokers (three of whom were above 45 years of age), there was no indication of genetic change.

As a result of public health campaigns in the United States, there has been a substantial reduction in the percentage of adults who smoke. From published figures, it can be estimated that there are approximately equal numbers of smokers and former smokers nationwide (about 43 million in each category) (33). At the University of Texas M. D. Anderson Cancer Center, Houston, more than half of the recently diagnosed lung cancers arise in former smokers, and nearly 50% of these had quit smoking more than 5 years previously (34). From the current smoking trends, it appears that former smokers will account for a growing percentage of all patients with lung cancer. Multiple mutations are found in invasive lung tumors, and, presumably, the diminished risk of former smokers is due to a decrease in the accumulation of new mutations in the bronchial epithelium. Somewhat surprisingly, we found no statistically significant differences in the frequencies or patterns of allelic loss between current and former smokers, and multiple molecular abnormalities were found in biopsy specimens from subjects who had quit 10–48 years previously. These findings suggest that molecular changes, unlike histologic changes, may persist long after smoking cessation. Our findings are consistent with the maintained increased risk of lung cancer in former smokers (7).

Subjects with previous aerodigestive cancers (including those of the lung) are at greatly increased risk for the development of second cancers (35–38). Thus, another unexpected finding was the lack of significant differences in molecular changes between subjects with and without a previous history of lung cancer.

The development of epithelial cancers requires multiple mutations (39), the step-wise accumulation of which may indicate a mutator phenotype (40,41). Thus, it is possible that those preneoplastic lesions that have accumulated multiple mutations are at highest risk for progression to invasive cancer. We found that 12% of histologically normal biopsy specimens had allelic loss equal to or greater than that present in CIS lesions. These findings suggest that CIS and invasive tumors may arise directly either from normal epithelium or from abnormal epithelium, without passing through the entire histologic sequence (parallel theory of cancer development) (42). In contrast, three CIS lesions lacked allelic loss in any of the regions studied. While no published data exist for CIS of the respiratory tract, multiple studies of the natural history of uterine cervical CIS indicate that only a subset progresses to invasive cancer [reviewed in (43)]. Our findings suggest that CIS and other histologically normal or abnormal foci having multiple regions of allelic loss are at increased risk of progressing to invasive cancer.

In colon cancer, the accumulation of mutations during the preneoplastic process is not random but usually follows a pattern (44). Similarly, in lung cancer, the developmental sequence is not random, with LOH at one or more chromosome 3p regions and at chromosome 9p21 being early events and RAS mutations occurring relatively late (12–15). Our finding that losses at chromosomes 3p and 9p are frequent and early events in current and former smokers without lung cancer are consistent with these observations. In contrast, LOH at chromosome 5q, a relatively frequent event in invasive lung cancer (45,46), was detected only in the CIS stage. The short arm of chromosome 3 (3p) contains several discrete regions, including 3p12, 3p14, 3p21, and 3p24–25, which are deleted in lung and other cancers and which are each believed to harbor recessive oncogenes (11). Deletions at one or more of these sites were frequently detected in histologically normal or slightly abnormal bronchial biopsy specimens.

We have previously demonstrated in patients with lung cancer that widely separated regions of the bronchial mucosa may demonstrate loss of the same allele of a polymorphic marker (13,14), a phenomenon that we have referred to as allele-specific loss (ASL). In this study, ASL was noted in 93% of paired comparisons. According to the cumulative binomial test, the probability of this finding happening by chance is 1 × 10−25. While ASL may represent clonality, we have suggested that alternate explanations may exist (13).

Alterations in microsatellite size are another genetic change associated with many cancers, including lung cancers (19,21–23). The relationship between microsatellite alterations and DNA repair mechanisms in lung cancer has not been established, but these alterations probably represent evidence of some form of genomic instability (24). Nevertheless, micro-satellite alterations are attractive candidates for the early molecular detection of cancer (19,20). We found microsatellite alterations in the normal and abnormal epithelia of smokers but not in non-smokers. Unlike LOH, the frequency of microsatellite alterations did not increase with more advanced histologic categories.

While this study was under review, a similar study was reported (25), describing frequent deletions in bronchial biopsy specimens from current and former smokers. Our results, which are in agreement with these findings, indicate that multiple clonal outgrowths of molecularly altered cells are widely distributed in the bronchial epithelium of smokers and that they persist for many years after smoking cessation. Our findings suggest the hypothesis that identifying biopsy specimens with extensive or certain patterns of allelic loss may provide new methods for assessing the risk of developing invasive lung cancer in smokers and for monitoring their response to chemoprevention. As with all diagnostic tests, these concepts will need to be validated in clinical trials.

Acknowledgments

Supported by Public Health Service contract N01CN45580-01 and grant 1P50CA70907–01 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services.

We thank Jing Xu for assistance with the statistical analysis.

References

  • 1.Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1996. CA Cancer J Clin. 1996;46:5–27. doi: 10.3322/canjclin.46.1.5. [DOI] [PubMed] [Google Scholar]
  • 2.Parkin DM, Pisani P, Lopez AD, Masuyer E. At least one in seven cases of cancer is caused by smoking. Global estimates for 1985. Int J Cancer. 1994;5904:494–504. doi: 10.1002/ijc.2910590411. [DOI] [PubMed] [Google Scholar]
  • 3.Saccomanno G, Archer VE, Auerbach O, Saunders RP, Brennan LM. Development of carcinoma of the lung as reflected in exfoliated cells. Cancer. 1974;33:256–270. doi: 10.1002/1097-0142(197401)33:1<256::aid-cncr2820330139>3.0.co;2-g. [DOI] [PubMed] [Google Scholar]
  • 4.Auerbach O, Stout AP, Hammond EC, Garfinkel L. Changes in bronchial epithelium in relation to smoking and cancer of the lung. N Engl J Med. 1961;265:253–267. doi: 10.1056/NEJM196108102650601. [DOI] [PubMed] [Google Scholar]
  • 5.Auerbach O, Hammond EC, Garfinkel L. Changes in bronchial epithelium in relation to cigarette smoking, 1955–1960 vs. 1970–1977. N Engl J Med. 1979;300:381–385. doi: 10.1056/NEJM197902223000801. [DOI] [PubMed] [Google Scholar]
  • 6.Bertram JF, Rogers AW. Recovery of bronchial epithelium on stopping smoking. Br Med J (Clin Res Ed) 1981;283:1567–1569. doi: 10.1136/bmj.283.6306.1567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.U.S. Department of Health and Human Service. The health benefits of smoking cessation. Washington (DC: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 1990. [Google Scholar]
  • 8.Schantz SP. Chemoprevention strategies: the relevance of premalignant and malignant lesions of the upper aerodigestive tract. J Cell Biochem Suppl. 1993;17F:18–26. doi: 10.1002/jcb.240531004. [DOI] [PubMed] [Google Scholar]
  • 9.Boone CW, Kelloff GJ. Intraepithelial neoplasia, surrogate endpoint biomarkers, and cancer chemoprevention. J Cell Biochem Suppl. 1993;17F:37–48. doi: 10.1002/jcb.240531007. [DOI] [PubMed] [Google Scholar]
  • 10.Gazdar AF, Bader S, Hung J, Kishimoto Y, Sekido Y, Sugio K, et al. Molecular genetic changes found in human lung cancer and its precursor lesions. Cold Spring Harb Symp Quant Biol. 1994;59:565–572. doi: 10.1101/sqb.1994.059.01.063. [DOI] [PubMed] [Google Scholar]
  • 11.Minna JD, Sekido Y, Fong K, Gazdar AF. Molecular biology of lung cancer, chapt 30.1. In: DeVita VT Jr, Hellman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. 5th. Philadelphia: Lippincott; 1997. pp. 849–857. [Google Scholar]
  • 12.Sugio K, Kishimoto Y, Virmani AK, Hung JY, Gazdar AF. K-ras mutations are a relatively late event in the pathogenesis of lung carcinomas. Cancer Res. 1994;54:5811–5815. [PubMed] [Google Scholar]
  • 13.Hung J, Kishimoto Y, Sugio K, Virmani A, McIntire DD, Minna JD, et al. Allele-specific chromosome 3p deletions occur at an early stage in the pathogenesis of lung carcinoma [published erratum appears in JAMA 1995; 273:1908] JAMA. 1995;273:558–563. [PubMed] [Google Scholar]
  • 14.Kishimoto Y, Sugio K, Hung JY, Virmani AK, McIntire DD, Minna JD, et al. Allele-specific loss in chromosome 9p loci in preneoplastic lesions accompanying non-small-cell lung cancers. J Natl Cancer Inst. 1995;87:1224–1229. doi: 10.1093/jnci/87.16.1224. [DOI] [PubMed] [Google Scholar]
  • 15.Sundaresan V, Ganly P, Hasleton P, Rudd R, Sinha G, Bleehen NM, et al. p53 and chromosome 3 abnormalities, characteristic of malignant lung tumours, are detectable in preinvasive lesions of the bronchus. Oncogene. 1992;7:1989–1997. [PubMed] [Google Scholar]
  • 16.Thiberville L, Payne P, Vielkinds J, LeRiche J, Horsman D, Nouvet G, et al. Evidence of cumulative gene losses with progression of pre-malignant epithelial lesions to carcinoma of the bronchus. Cancer Res. 1995;55:5133–5139. [PubMed] [Google Scholar]
  • 17.Chung GT, Sundaresan V, Hasleton P, Rudd R, Taylor R, Rabbitts PH. Clonal evolution of lung tumors. Cancer Res. 1996;56:1609–1614. [PubMed] [Google Scholar]
  • 18.Smith AL, Hung J, Walker L, Rogers TE, Vuitch F, Lee E, et al. Extensive areas of aneuploidy are present in the respiratory epithelium of lung cancer patients. Br J Cancer. 1996;73:203–209. doi: 10.1038/bjc.1996.36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Mao L, Lee DJ, Tockman MS, Erozan YS, Askin F, Sidransky D. Microsatellite alterations as clonal markers for the detection of human cancer. Proc Natl Acad Sci U S A. 1994;91:9871–9875. doi: 10.1073/pnas.91.21.9871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Miozzo M, Sozzi G, Musso K, Pilotti S, Incarbone M, Pastorino U, et al. Microsatellite alterations in bronchial and sputum specimens of lung cancer patients. Cancer Res. 1996;56:2285–2288. [PubMed] [Google Scholar]
  • 21.Fong KM, Zimmerman PV, Smith PJ. Micro-satellite instability and other molecular abnormalities in non-small cell lung cancer. Cancer Res. 1995;55:28–30. [PubMed] [Google Scholar]
  • 22.Adachi J, Shiseki M, Okazaki T, Ishimaru G, Noguchi M, Hirohashi S, et al. Microsatellite instability in primary and metastatic lung carcinomas. Genes Chromosomes Cancer. 1995;14:301–306. doi: 10.1002/gcc.2870140409. [DOI] [PubMed] [Google Scholar]
  • 23.Merlo A, Mabry M, Gabrielson E, Vollmer R, Baylin SB, Sidransky D. Frequent microsatellite instability in primary small cell lung cancer. Cancer Res. 1994;54:2098–2101. [PubMed] [Google Scholar]
  • 24.Fishel R, Lescoe MK, Rao MR, Copeland NG, Jenkins NA, Garber J, et al. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer [published erratum appears in Cell 1994;77: 167] Cell. 1993;75:1027–1038. doi: 10.1016/0092-8674(93)90546-3. [DOI] [PubMed] [Google Scholar]
  • 25.Mao L, Lee JS, Kurie JM, Fan YH, Lippman SM, Lee JJ, et al. Clonal genetic alterations in the lungs of current and former smokers. J Natl Cancer Inst. 1997;89:857–862. doi: 10.1093/jnci/89.12.857. [DOI] [PubMed] [Google Scholar]
  • 26.Lam S, MacAulay C, Hung J, LeRiche J, Profio AE, Palcic B, et al. Detection of dysplasia and carcinoma in situ with a lung imaging fluorescence endoscope device. J Thorac Cardiovasc Surg. 1993;105:1035–1040. [PubMed] [Google Scholar]
  • 27.Lam S, MacAulay C, LeRiche JC, Ikeda N, Palcic B. Early localization of bronchogenic carcinoma. Diagnos Therapeut Endos. 1994;1:75–78. doi: 10.1155/DTE.1.75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Cawkwell L, Lewis FA, Quirke P. Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction. Br J Cancer. 1994;70:813–818. doi: 10.1038/bjc.1994.404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Jones MH, Nakamura Y. Detection of loss of heterozygosity at the human TP53 locus using a dinucleotide repeat polymorphism. Genes Chromosomes Cancer. 1992;5:89–90. doi: 10.1002/gcc.2870050113. [DOI] [PubMed] [Google Scholar]
  • 30.Toguchida J, McGee TL, Paterson JC, Eagle JR, Tucker S, Yandell DW, et al. Complete genomic sequence of the human retinoblastoma susceptibility gene. Genomics. 1993;17:535–543. doi: 10.1006/geno.1993.1368. [DOI] [PubMed] [Google Scholar]
  • 31.Armitage P, Berry G. Statistical methods in medical research. Palo Alto (CA): Blackwell Scientific Publications; 1987. [Google Scholar]
  • 32.Siegel S. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill; 1956. [Google Scholar]
  • 33.Morbidity and Mortality Weekly Report. Vol. 43. Anon: Office of Smoking and Health, U.S. Centers for Disease Control; 1994. [Google Scholar]
  • 34.Tong L, Spitz MR, Fueger JJ, Amos CA. Lung carcinoma in former smokers. Cancer. 1996;78:1004–1010. doi: 10.1002/(SICI)1097-0142(19960901)78:5<1004::AID-CNCR10>3.0.CO;2-6. [DOI] [PubMed] [Google Scholar]
  • 35.Slaughter DP, Southwick HW, Smejkal W. “Field cancerization” in oral stratified squamous epithelium: clinical implications of multicentric origin. Cancer. 1954;6:963–968. doi: 10.1002/1097-0142(195309)6:5<963::aid-cncr2820060515>3.0.co;2-q. [DOI] [PubMed] [Google Scholar]
  • 36.Heyne KH, Lippman SM, Lee JJ, Lee JS, Hong WK. The incidence of second primary tumors in long-term survivors of small-cell lung cancer. J Clin Oncol. 1992;10:1519–1524. doi: 10.1200/JCO.1992.10.10.1519. [DOI] [PubMed] [Google Scholar]
  • 37.Noguchi M, Maezawa N, Nakanishi Y, Matsuno Y, Shimosato Y, Hirohashi S. Application of the p53 gene mutation pattern for differential diagnosis of primary versus metastatic lung carcinomas. Diagn Mol Pathol. 1993;2:29–35. [PubMed] [Google Scholar]
  • 38.Johnson BE, Linnoila RI, Williams JP, Venzon DJ, Okunieff P, Anderson GB, et al. Risk of second aerodigestive cancers increases in patients who survive free of small-cell lung cancer for more than 2 years. J Clin Oncol. 1995;13:101–111. doi: 10.1200/JCO.1995.13.1.101. [DOI] [PubMed] [Google Scholar]
  • 39.Fisher JC. Multiple mutation theory of carcinogenesis. Nature. 1958;181:651–652. doi: 10.1038/181651b0. [DOI] [PubMed] [Google Scholar]
  • 40.Loeb LA. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res. 1991;51:3075–3079. [PubMed] [Google Scholar]
  • 41.Loeb LA. Microsatellite instability: marker of a mutator phenotype in cancer. Cancer Res. 1994;54:5059–5063. [PubMed] [Google Scholar]
  • 42.O’Connell P, Pekkel V, Fuqua S, Osborne CK, Allred DC. Molecular genetic studies of early breast cancer evolution. Breast Cancer Res Treat. 1994;32:5–12. doi: 10.1007/BF00666201. [DOI] [PubMed] [Google Scholar]
  • 43.Wright TC, Kurman RJ, Ferenczy A. Precancerous lesions of the cervix. In: Kurman RJ, editor. Blaustein’s pathology of the female genital tract. New York: Springer-Verlag; 1994. pp. 229–277. [Google Scholar]
  • 44.Fearon ER. Molecular genetic studies of the adenoma-carcinoma sequence. Adv Intern Med. 1994;39:123–147. [PubMed] [Google Scholar]
  • 45.D’Amico D, Carbone DP, Johnson BE, Meltzer SJ, Minna JD. Polymorphic sites within the MCC and APC loci reveal very frequent loss of heterozygosity in human small cell lung cancer. Cancer Res. 1992;52:1996–1999. [PubMed] [Google Scholar]
  • 46.Wieland I, Bohm M. Frequent allelic deletion at a novel locus on chromosome 5 in human lung cancer. Cancer Res. 1994;54:1772–1774. [PubMed] [Google Scholar]

RESOURCES