Lung Cancer (2005) 47, 205—214 Lung carcinoma-associated atypical adenomatoid hyperplasia, squamous cell dysplasia, and chromosome alterations in non-neoplastic bronchial mucosa Klaus Kaysera,∗, Zdravko Kosjerinab, Torsten Goldmannc, Gian Kayserd, Bernd Kazmierczake, Ekkehard Vollmere a UICC-TPCC, Institute of Pathology, Charite, Schumann Str. 20/21, D-10168 Berlin, Germany Department of Pathology, University Nova Sad, Serbia c Clinical and Experimental Pathology, Research Center Borstel, Borstel, Germany d Institute of Pathology, University Freiburg, Freiburg, Germany e Institute of Molecular Genetics, Bremen, Germany b Received 29 March 2004 ; received in revised form 23 June 2004; accepted 2 July 2004 KEYWORDS Lung carcinoma; Atypical adenomatoid hyperplasia; Squamous cell dysplasia; Karyotype analysis Summary This article analyzes phenotype and genotype alterations of the lung in association with lung cancer. The frequency of phenotype preneoplastic lesions (atypical adenomatoid hyperplasia (AAH) and squamous cell dysplasia (SCD)) was analyzed at distinct distances from the tumor boundary in 150 lung carcinomas. AAH was noted in 19/150 (13%) cases and more frequently seen in adeno carcinomas, squamous cell dysplasia was noted in 46/150 (31%) cases and more frequently seen in squamous cell carcinomas. The degree of cellular atypia decreased with increasing distance from tumor boundary in both AAH and SCD. At similar distances, genotype (chromosome) alterations of surrounding bronchial mucosa were studied in additional 55 primary and secondary lung tumors by karyotype analysis. Numerical chromosome aberrations occur frequently in primary lung carcinomas and adjacent bronchial mucosa, and affect at average 4.5/10 metaphases in primary lung cancer and 2/10 metaphases in metastases. Most abnormal metaphases were induced by chromosome losses, only few by additional copies, i.e. trisomy, etc. Losses of y chromosome were seen in both malignancy and adjacent bronchial mucosa, and interpreted as ‘‘tumor related’’, losses of chromosome 21 in adjacent bronchial mucosa were non-tumor related in * Corresponding author. Tel.: +49 62 21413827; fax: +49 62 21451570. E-mail address: Klaus.Kayser@charite.de (K. Kayser). 0169-5002/$ — see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2004.07.042 206 K. Kayser et al. adenocarcinoma and metastases, losses of chromosome 19 in adjacent bronchial mucosa occurred independently in squamous cell and large cell carcinomas. The data suggest the hypothesis that preneoplastic lesions in the lung might be partly induced by the tumor itself. © 2004 Elsevier Ireland Ltd. All rights reserved. 1. Introduction The majority of scientists are convinced that the development of cancer, especially lung cancer occurs in multiple distinct steps, which can be clearly distinguished from each other by phenotype and genotype examinations [1,2]. These lesions are commonly named pre-neoplasia, and morphologically form a constant ‘‘ladder’’ starting from ‘‘mild atypia’’ via carcinoma in situ to fully developed cancer [3—11]. The lung is the organ most intensively exposed to environmental potential hazardous substances, and, being this, these different phenotype steps should be clearly recognizable and associated with lung carcinomas. Two major lesions have now been recognized to be ‘‘pre-cursor stages’’ of lung carcinomas: squamous cell dysplasia (SCD) of bronchial mucosa, and atypical adenomatoid hyperplasia (AHH) of peripheral (alveolar) lung parenchyma [1,2,12]. SCD is well known, and a common finding in patients with chronic inflammatory lesions of the conducting air systems, such as heavy smokers, chronic obstructive lung diseases, or asthma [1,13]. SCD has been reported to be reversible even at the stage of carcinoma in situ [13]. If SCD is indeed a pre-neoplastic lesion, it should be associated with already existent lung carcinomas in terms of cell type and spatial relationship [13]. To our knowledge only one detailed study has been published which analyzed the frequency of SCD in relation to its distance from the boundary of lung carcinomas [13]. This singular study confirmed an association of SCD with squamous cell carcinomas and a decreasing frequency of this lesion with increasing distance from tumor boundary. AAH is a less known small lesion of peripheral lung epithelial cells, measuring about 5 mm in maximum diameter, and characterized by sheets of proliferating alveolar lining and cuboidal cells with dense nuclear chromatin [12—17]. This focal lesion of the peripheral lung, which can already be detected by high-resolution CT-techniques [14], has striking similarities with well-differentiated adenocarcinomas of the bronchiolo-alveolar type. It possesses well-defined boundaries, slightly thickened inter-alveolar septa, infrequently bronchiolar metaplasia, and is often observed at the boundary of adenocarcinomas of the lung [18,19]. The frequency of these lesions has been reported to range from 5 to 25% in resected lung specimens [20—22]. Cytometric studies supported by profiling of growth-related markers showed that AAH is often characterized by non-diploid cellular proliferation which can even be of monoclonal origin [23,24]. At present, it is of specific interest to pathologists and epidemiologists as the incidence of peripheral adenocarcinomas of the lung is increasing, and AAH is considered to be closely associated with the development of peripheral adenocarcinomas. In addition to phenotype investigations, genotype studies on lung cancer revealed numerous abnormalities, mainly detected on chromosome 3, 17, 19, 21 [25—27]. These findings fit to the well-known phenotype heterogeneity of lung cancer [9,28—31], which has been also measured with static and flow cytometric DNA techniques. All these studies inform about phenotype and genotype specificities of lung cancer or SCD, only one study on genotype of AAH [27], none of these studies has analyzed the association of phenotpye and genotype alterations of nonneoplastic lung parenchyma with that of resected lung carcinomas. In extending the known separate information on lung carcinomas or pre-neoplastic lung lesions, this study focuses on both phenotype detection of SCD and AAH of peripheral lung, and genotype classification of tumor-associated non-neoplastic bronchial mucosa in terms of chromosome analysis (losses and gains). Thus, it should clarify questions such as: (a) are pre-neoplastic lesions associated with primary lung carcinomas; (b) do intra-pulmonary malignancies influence the phenotype and genotype of nonneoplastic lung parenchyma? 2. Materials and methods This study includes 205 lobes and lungs of potentially curative operated lung carcinoma patients sent to the Institute of Pathology, Research Center Borstel, Borstel, Germany. The specimens were macroscopically and microscopically analyzed by three of the authors (K.K., Z.K., E.V.), and a consensus diagnosis was given based upon conventional HE and PAS stains as well as on immunohistochemistry (keratin, vimentin, or Preneoplasia related to Lung carcinoma 207 neuroendocrine markers), if needed. A total of 150 surgical excisions underwent a thorough search for pre-neoplastic lesions of bronchi and peripheral lung parenchyma as follows: The specimens were fixed with buffered formalin via the bronchi, cut into serial sections 5 mm thick, and four bronchial and four tissue sections at each distances 1, 3, 5, 7, 9 cm from the tumor boundary were analyzed for SCD and AHH, respectively. Immediately after excision, biopsies were taken from further 55 non-fixed surgical specimens at the tumor boundary, and at distances 1—4 cm from the tumor boundary. These biopsies measuring approximately 5 mm × 5 mm × 5 mm were immediately placed into tissue culture medium and prepared for cytogenetic analysis, which was performed and karyotyped by the same cytogeneticist (B.K.) in all cases. The remaining tissue was fixed with buffered formalin, and subject for histologic examinations as usual. From each tumor four representative tissue blocks (two from the center and two from the boundary) were stained with HE, PAS, and immunohistochemical markers (keratin, vimentin, synaptophysin, etc.), if necessary. Conventional chromosomal analysis of the tumors and the non-tumorous bronchial biopsies was performed after short-term culture by Giemsa (G)banding. From each probe 10 metaphases were analyzed, karyotyped, and the numerical aberrations (losses, additional copies such as triplets, etc.) were counted. For each probe 10 metaphases were analyzed. A case was graded as abnormal, if at least one out of the 10 metaphases revealed numerical alterations, either losses or additional chromosome copies. The technical procedure has been described in detail elsewhere [32,33]. The final statistical assessments were performed using a commercially available program pack- Table 1 age (NCSS, Number Cruncher Statistical System, Kaysville, USA), which included chi-square, f-, and non-parametric tests as well as multivariate discriminate and regression analysis. 3. Results A synopsis of the prospective phenotype material is given in Table 1. It includes 75 cases which have contributed to a previous study [12]. The age distribution of the patients, sex ratio, and frequency of pT and pN stages does not differ to previous studies [34—36]. Women are insignificantly younger than men at date of operation; advanced tumor stages contributed to 27% (pT3, pT4) and to 30% (pN2, pN3), respectively. In 22/150 cases, AAH could be detected, which was more frequently seen in adeno carcinomas compared to squamous cell carcinomas (P < 0.05). High grade AAH was found in 15 (68.2%) cases and low grade AAH in seven (31.8%) cases. AAH is not infrequently a multiple lesion, as multiple lesions could be detected in six cases with a maximum of seven clearly separated lesions in a case with an adeno carcinoma. The frequency distribution of AAH in relation to the distance from the tumor boundary is depicted in Fig. 1. About one-half of the lesions were seen within a distance of 3 cm to the tumor boundary, whereas nine lesions have been noted at a distance >7 cm. There was no preference to a right or left side or the upper and lower lobes of the lungs. The degree of atypia seems to decrease with increasing distance to the tumor boundary. The data of SCD differ from those of AAH, as shown in Table 2. In 38.6% of cases with squamous cell carcinomas a SCD could be detected in con- Age distribution and frequency of atypical adenomatoid hyperplasia (AAH) Feature Number of cases Age (mean + S.D.) Men Women 132 18 62 ± 8.5 59 ± 12.3 Total 150 61.7 ± 9.0 With AAH Without AAH 22 128 64 ± 6.7 61.5 ± 9.1 Cell type Squamous Adeno Large cell Others/benign Total Cases with AAH Percent with AAH (%) 7/70 10/54 1/7 4/19 10 18.5 14.2 21.0 22/150 14.7 208 K. Kayser et al. Table 3 Cytogenetic study (55 cases), clinical data and cell type Fig. 1. Number of detected AAH in relation to distance from tumor boundary. Table 2 Frequency and localization of SCD Cell type Cases with SCD Percent with SCD (%) Squamous Adeno Large cell Others/benign 27/70 11/54 2/7 6/19 38.6 20.3 28.5 31.5 Total 46/150 30.7 trast to only 20.3% of adeno carcinomas. Thus, SCD is more frequent compared to AAH and is closely associated with a different tumor cell type. The relationship of SCD with the distance from the tumor boundary is demonstrated in Fig. 2. Most of the lesions were seen at a distance 1—3 cm from the tumor boundary. This finding is independent from the tumor cell type. Again, SCD is a multiple lesion, and 2.1 lesions were seen at average in a lobe/lung. The synopsis of patients included into the chromosome study is given in Table 3. Patients with Cell type Men Women Total Squamous Adeno Large cell Metastasis Mesothelioma Carcinoids 17 13 4 3 1 1 3 11 0 2 0 0 20 24 4 5 1 1 Total 39 16 55 Age 61.2 ± 9.1 59.3 ± 10.6 60.6 ± 9.7 adeno carcinomas contribute to the majority of cases followed from those with squamous cell carcinomas. A characteristic chromosome pattern showing a loss of the y chromosome in the tumor and at a distance of 3 cm in a male patient is depicted in Fig. 3. The analysis of the data focuses on: (a) the relative frequency of cases which display at least one numerical chromosome alteration; and (b) the number of abnormal metaphases (losses or additional chromosome copies) within the analyzed 10 metaphases. (a) The relative frequency of chromosome losses in the analyzed tumors and adjacent bronchial mucosa is shown in Table 4. Losses of chromosome y were noted in 56% of cases in the tumors and of similar frequency in the adjacent bronchus mucosa. (b) The number of abnormal metaphases in relation to the measured 10 metaphases was the lowest in the carcinoid (1.5 ± 1.0), metastases and the mesothelioma formed the second group presenting with 2.1 ± 1.5 numerical aberrations at average. Basically, primary lung car- Fig. 2. Number of detected SCD in relationship to cell type and distance from tumor boundary. Preneoplasia related to Lung carcinoma 209 Fig. 3. Karyotype of tumor and adjacent bronchial mucosa showing a loss of chromosome y in the tumor and at a distance of 3 cm. cinomas formed one singular group presenting with 4.5 ± 3.6 numerical aberrations at average. These numbers were also seen in the adjacent bronchial mucosa, independently from tumor cell type. They decreased slightly with increasing distance from the tumor boundary as demonstrated in Fig. 4. Within the metaphases, loss of the y chromosome was most frequently observed in our material (men only). It decreases with increasing distance, especially at a distance 4 cm (not shown). The most frequent numerical chromosome aberrations in relation to cell type and distance are presented in Table 5. The losses of chromosome y can be seen in most tumors and in the adjacent bronchial mucosa in contrast to the losses of chromosome 19, which were not seen in squamous cell carcinomas, however frequently in their tumor-free adjacent bronchial mucosa. The statistical evaluation of the relationship of numerical chromosome aberrations between the Table 4 Chromosome losses of tumors and non-tumorous bronchial mucosa in relation to its distance of tumor boundary Tissue Chromosome (N = 55) −y Tumors 56 Bronchus mucosa 1 cm 68 2 cm 61 3 cm 58 4 cm 35 −8 −15 −19 −21 −22 −4* −11* −17* 4 4 18 16 16 4 8 16 5 7 5 6 16 9 12 10 23 24 25 24 18 18 13 9 7 20 15 13 10 6 6 6 12 8 4 2 8 16 16 16 In percent of cases; (*) structural aberrations of these chromosomes have been reported to be associated with exogenous noxes in inbred mice [42]. 210 K. Kayser et al. Fig. 4. Number of chromosome losses in different tumors and adjacent bronchial mucosa, derived from analysis of 10 metaphases. tumor and the tumor-free adjacent bronchial mucosa results in data presented in Table 6. Losses of chromosome 21 were seen contemporary in tumorfree bronchial mucosa and in squamous cell and large cell carcinomas only, and were interpreted to be tumor-associated. To the contrast, these losses could not be observed in adeno carcinomas and metastases, and were thus interpreted to be tumor independent for adeno carcinomas and metastases. The opposite holds true for losses of chromosome 19, which were tumor-independent for squamous cell and large cell carcinomas, and tumor dependent for adeno carcinomas. 4. Discussion The still high incidence of bronchial carcinomas and the high percentage of advanced tumor stages at date of clinical manifestation require new strategies of lung cancer detection and prevention [1,34]. Although the main risk factors for lung cancer development such as heavy smoking or asbestos exposure are well known, prevention studies on populations at high risk have failed until now to our knowledge [1,37,38]. One reason of failure could be seen in the ‘‘soft’’ definition of so-called preneoplastic lesions of bronchial carcinomas: In contrast to the well established squamous metaplasia and dysplasia of the cervix and clear-cut therapeutic advices how to treat these lesions [3,39,40], the situation of SCD in the bronchi is less well examined and probably more complicated. First, the air con- ducting system of the lung can be only inspected in its proximal areas, second it might contemporary express several pre-neoplastic lesions, and third, it is not the only phenotype lesion of the lung which is suspicious for a pre-cancerous stage [1]. From the phenotype point of view, at least two pre-neoplastic lesions can be distinguished: SCD and AAH [9,27,30,41]. Both lesions are morphologically well defined, however, most pathologist do not care about these lesions, as they are normally only insignificant findings compared to the contemporary excised lung cancer. Detailed examinations on AAH could demonstrate that these lesions are of prognostic value: patients who developed AAH in addition to their surgically treated lung carcinoma presented with poorer survival in comparison to those without AAH [2,12,27,28]. The frequency of these lesions (SCD and AAH) has been reported to range from 5 to 25% in resected lung specimens [20—22]. Population-based incidence studies are still missing to our knowledge. This study confirms the data of the literature [2,12]: although the evaluation of AAH frequency in resected lung specimens is obviously closely associated with the minuteness of detail its frequency probably exceeds 15%. In addition, it is a multiple lesion closely associated with the tumor cell type: It can be noted about twice as often in adeno carcinoma than in squamous cell carcinomas. Using these results for ‘‘internal control’’, the analysis of spatial relationship of AAH to the tumor boundary implies more details about this lesion: (A) it is most frequently seen at a distance 1—3 cm from the tumor boundary. (B) AAH with high-grade cellular 17 10 7, 18 5, 20 211 Table 6 Relationship of chromosome losses between tumor tissue and adjacent bronchial mucosa Cell type Tumorrelated Not related to tumor Squamous Large cell Adeno Metastases Others y, 21, 22 y, 21 y, 19, 22 y y 19 15, 19 15, 21 x, 21 8, 21 7, 4, 6, 3, 7, 8, 21 17, 18 6 6, 18, 19 3, 4, 5, 3, 7, 20, 22 3, 7, 8, 9, 17 5, 7, 18 12, 18, 20, 22 19 4, 10, 22 5, 9, 18 3, 4, 5, 20, 21, 22 17 4, 7 9 17 7, 3, 5, 7, 4, 2, 4, 2, y, 7, 19 y, 15, 21 y, 15, 17, 19 x, y, 21 19, 22 15, 19, 21, 22 19, 15, 21 8, 15, 17, 20, 22 y, y, y, y, y, 6, 16, 17, 18, 19, 22 y, 17, 18, 19, 21, 22 y, 6, 15, 19, 21 x, 13, 19, 21 y, 9, 15, 19, 21 y, 5, 13, 15, 19, 22 y, 10, 19, 21 x, y, 3, 4, 8, 9, 10, 11, 20, 21, 22 16, 18, 20, 21, 22 9, 17, 18, 19, 22 10, 12, 17, 21 6, 12, 18, 19 4 cm 3 cm 2 cm 1 cm Tumor y, y, y, y, Losses Squamous Adeno Large cell Metastases Additonal copies Squamous Adeno Large cell Metastases Cell type Table 5 Most frequently observed chromosome losses and additional chromosome copies in lung cancer, metastases, and tumor-free adjacent bronchial mucosa (cut-off level 10% of cases) Preneoplasia related to Lung carcinoma atypia is closer associated to the tumor boundary than those lesions with low grade AAH. (C) In benign lesions and metastases AAH is not associated to the distance from these lesions. When comparing the results of AAH with those of SCD the frequency of SCD exceeds that of AAH for a factor 2. In regard to the other features the similarities between AAH and SCD are striking: (a) both are a multiple lesion, and at average 2.2 lesions exist; (b) SCD is seen most frequently in its derived tumor cell type too, i.e., squamous cell carcinomas; (c) its most frequent localization from primary lung carcinomas is at the same distance as AAH, i.e., at 1—3 cm; (d) its localization is not associated from the lesions’ boundary for metastases and benign lesions (Fig. 2). These findings confirm the data of a previous study on SCD: The occurrence and localization of squamous cell dysplasia is closely associated with squamous cell carcinoma, still, however with less significance to the other cell types of primary lung carcinomas, and not related to metastases [2,12]. To interpret these data, the assumption that SCD and AAH are only pre-neoplastic lesions, i.e., lesions which precede the establishment of cancer, seems not to be sufficient: the pre-neoplastic lesions which has induced the carcinoma should already be overgrown by the carcinoma itself, and no longer be visible. The localization of additional existing lesions should not be related to that of the primary lung cancer. Thus, it seems reasonable to assume that the carcinoma itself might also induce these lesions. Is this hypothesis supported by our genotype data? When looking to genotype studies of preneoplastic lesions, cytometric studies supported by profiling of growth-related markers showed that AAH is often characterized by non-diploid cellular proliferation which can even be of monoclonal origin [23,24]. In respect to genetic instability induced by air pollution, linkage analyses using inbred mice identified chromosomal segments (quantitative trait loci (QTL)), with genes controlling the 212 susceptibility of the lung to inflammation (chromosome 17), injury (chromosome 11), and hyperpermeability (chromosome 4) in responses to ozone (03) exposure [42]. In our study, numerical aberrations of these chromosomes were also noted in both, tumor tissue and bronchial mucosa. The percentage of cases displaying losses of chromosome 4 and 17 (hyperpermeability and inflammation) displayed no relation to distance from the tumor boundary in contrast to that of chromosome 11 (injury), which decreased with increasing distance (Table 4). In relation to phenotype analysis, Ullmann et al. (2003) described a new pre-neoplastic lesion in the peripheral lung parenchyma, which they called bronchiolar columnar cell dysplasia [27]. By use of comparative genomic hybridization (CGH) they found genetic aberrations in 5/6 cases, mainly losses of chromosomes 3, 9, 10, 13, 14, and numerical increase of chromosomes 1,17, 19, 20 [27], i.e., chromosome alterations in peripheral lung parenchyma are not only associated to ‘‘common’’ AAH. In addition, AAH is characterized by specific alterations of carbohydrate binding capacities, expression of galectins, and of calcyclin, a protein of the S100 family [2]. Irreversible alterations of vascularization form additional characteristics [12,43]. All these patterns should reflect in alterations of the genotype. In contrast to previous studies on phenotype examinations of AAH [16,44—46] the measurements of genotype in this study were performed with short-term tissue culture and karyotyping of chromosomes. This technique permits a fast and reliable analysis of numerical and structural chromosome aberrations [32,33]. In contrast to other genotype analysis techniques such as comparative genomic hybridization (CGH) or in situ hybridization, chromosomes are analyzed in a specific stage of the cell cycle (metaphases). The results reflect to chromosome abnormalities, which might be expressed during cell proliferation only. The disadvantage is the missing phenotype classification of the lesion. The results are surprising, as they display chromosome alterations in the tumor itself, which has been expected. In addition, frequent numerical chromosome aberrations in bronchial tissue macroscopically not being involved by tumor growth, and quite far away from tumor boundary were also discovered. Moreover, these alterations were seen more intensively in the bronchus mucosa than in tumor tissue of metastases, and display the same spatial association as found for pre-neoplasia described by phenotype. When looking specifically to the involved chromosomes, two sets of alterations can be distinguished: (a) those which are also present in the tumor tissue, and which have K. Kayser et al. been interpreted to be tumor-associated, and (b) those, which are only present in the bronchial mucosa, independently from its distance, and which suggest to be of tumor-independent nature. Tumor related are losses of chromosome y in all tumor cell types, those of chromosome 21, 22 in squamous cell carcinomas, and those of chromosome 19, 22 in adeno carcinomas (Table 4). The corresponding gains are those of chromosome 7 and 4, respectively. These findings fit into the theory, that genotype alterations of cancer-inducing lesions will be maintained in the derived cancer, whereas the additional observation, that losses of certain chromosomes can be seen in the adjacent tumor free tissue, but not in the tumor itself, require a different explanation: They reflect either to a preexisting genetic instability, or they are induced by external factors such as inhaled toxic substances or by ‘‘indirect’’ tumor influence. Chromosomes 11 and 17 are subject to external damage by inflammation and injury [42], which seem to be not involved according to our study. The distinct association of these chromosomes to the tumor cell type (chromosome 19 in squamous cell carcinoma, and chromosome 15, 21 in adeno carcinoma only) suggest that these damages might be induced by the tumor itself. Thus, these data, which are basically in agreement with the so-called field theory of cancer manifestation, fits into the observation of phenotype: So-called pre-neoplastic lesions of the lung in terms of SCD and AAH might be not always pre-existing lesions prior to tumor manifestation. They might, in addition, also induced by the tumor growth itself, thus refer to the so-called malignancy associated changes [1]. Furthermore, numerical chromosome aberrations measured by karyotyping frequently occur in resected lung specimens with lung cancer. As these alterations can be observed still in 35% of cases in bronchial mucosa at a distance of 4 cm from the tumor boundary and occur in about 40% of dividing cells, they indicate genetic instability and might explain several observations such as focal manifestation of cancer, pre-neoplasia, or malignancy associated changes, and should be subject of further investigations. Acknowledgment The financial support of the International Association for the Study of Lung Cancer (IASLC) and the Verein zur Forderung des Biologisch Technologischen Fortschritts in der Medizin e.V. are gratefully acknowledged. Preneoplasia related to Lung carcinoma 213 References [1] Kayser K. Analytical Lung Pathology. New York: Heidelberg, Springer; 1992. [2] Kayser K, Andre S, Bovin NV, Zeng FY, Gabius HJ. Preneoplasia-associated expression of calcyclin and of binding sites for synthetic blood group A/H trisaccharideexposing neoglycoconjugates in human lung. Cancer Biochem Biophys 1997;15:235—43. [3] Koss LG. Significance of dysplasia. Clin Obstet Gynecol 1970;13:873—88. [4] Sugimachi K, Sumiyoshi K, Nozoe T, Yasuda M, Watanabe M, Kitamura K, et al. Carcinogenesis and histogenesis of esophageal carcinoma. Cancer 1995;75:1440—5. [5] Franklin WA, Folkvord JM, Varella-Garcia M, Kennedy T, Proudfoot S, Cook R, et al. Expansion of bronchial epithelial cell populations by in vitro culture of explants from dysplastic and histologically normal sites. Am J Respir Cell Mol Biol 1996;15:297—304. [6] Ren ZP, Hedrum A, Ponten F, Nister M, Ahmadian A, Lundeberg J, et al. Human epidermal cancer and accompanying precursors have identical p53 mutations different from p53 mutations in adjacent areas of clonally expanded nonneoplastic keratinocytes. Oncogene 1996;12:765—73. [7] Partridge M, Emilion G, Pateromichelakis S, Phillips E, Langdon J. Field cancerisation of the oral cavity: comparison of the spectrum of molecular alterations in cases presenting with both dysplastic and malignant lesions. Oral Oncol 1997;33:332—7. [8] Scholes AG, Woolgar JA, Boyle MA, Brown JS, Vaughan ED, Hart CA, et al. Synchronous oral carcinomas: independent or common clonal origin? Cancer Res 1998;58:2003—6. [9] Keith RL, Miller YE, Gemmill RM, Drabkin HA, Dempsey EC, Kennedy TC, et al. Angiogenic squamous dysplasia in bronchi of individuals at high risk for lung cancer. Clin Cancer Res 2000;6:1616—25. [10] Khaled A, Imamura Y, Noriki S, Fukuda M. Early progression stage of malignancy of uterine cervical dysplasia as revealed by immunohistochemical demonstration of increased DNA-instability. Eur J Histochem 2000;44:143—56. [11] Saito T, Sugiura C, Hirai A, Notani K, Totsuka Y, Shindoh M, et al. Development of squamous cell carcinoma from preexistent oral leukoplakia: with respect to treatment modality. Int J Oral Maxillofac Surg 2001;30:49—53. [12] Kayser K, Nwoye JO, Kosjerina Z, Goldmann T, Vollmer E, Kaltner H, et al. Atypical adenomatous hyperplasia of lung: its incidence and analysis of clinical, glycohistochemical and structural features including newly defined growth regulators and vascularization. Lung Cancer 2003;42: 171—82. [13] Kayser K, Hagemeyer O, Runtsch T. Morphologic lesions in non-neoplastic bronchial mucosa associated with bronchial carcinomas. Zentralbl Pathol 1991;137:425—9. [14] Kawakami S, Sone S, Takashima S, Li F, Yang ZG, Maruyama Y, et al. Atypical adenomatous hyperplasia of the lung: correlation between high-resolution CT findings and histopathologic features. Eur Radiol 2001;11:811—4. [15] Leslie KO, Colby TV. Pathology of lung cancer. Curr Opin Pulm Med 1997;3:252—6. [16] Mori M, Rao SK, Popper HH, Cagle PT, Fraire AE. Atypical adenomatous hyperplasia of the lung: a probable forerunner in the development of adenocarcinoma of the lung. Mod Pathol 2001;14:72—84. [17] Mori M, Tezuka F, Chiba R, Funae Y, Watanabe M, Nukiwa T, et al. Atypical adenomatous hyperplasia and adenocarcinoma of the human lung: their heterology in form and [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] analogy in immunohistochemical characteristics. Cancer 1996;77:665—74. Shimosato Y, Noguchi N, Matsuno Y. Adenocarcinoma of the lung: its development and malignant progression. Lung Cancer 1993;9:99—108. Carey FA, Fallace WA, Fergusoson RJ, Kerr KM. Alveolar atypical hyperplasia in association with primary pulmonary adenocarcinoma: a study of 10 cases. Thorax 1992;47:1041—43. Nakahara R, Yokose T, Nagai K, Nishiwaki Y, Ochiai A. Atypical adenomatous hyperplasia of the lung: a clinicopathological study of 118 cases including cases with multiple atypical adenomatous hyperplasia. Thorax 2001;56:302—5. Weng S, Tsuchiya E, Kasuga T, Sugano H. Incidence of atypical bronchioalveolar cell hyperplasia of the lung: relation to histological subtypes of lung cancer. Virchows Arch Pathol Anatom 1992;420:463—71. Kosjerina ZVE, Goldmann T, Kayser K. Frequency in surgical specimens and morphology of atypical alveolar hyperplasia of the lung. Elec J Pathol Histol 2002;8.3:023—004. Yokozaki M, Kodama T, Yokose T, Matsumoto T, Mukai K. Differentiation of atypical adenomatous hyperplasia and adenocarcinoma of the lung by use of DNA ploidy and morphometric analysis. Mod Pathol 1996;9:1156—64. Niho S, Yokose T, Suzuki K, Kodama T, Nishiwaki Y, Mukai K. Monoclonality of atypical adenomatous hyperplasia of the lung. Am J Pathol 1999;154:249—54. Kerr KM. Pulmonary preinvasive neoplasia. J Clin Pathol 2001;54:257—71. Sozzi G, Miozzo M, Donghi R, Pilotti S, Cariani CT, Pastorino U, et al. Deletions of 17p and p53 mutations in preneoplastic lesions of the lung. Cancer Res 1992;52:6079—82. Ullmann R, Bongiovanni M, Halbwedl I, Petzmann S, GoggKammerer M, Sapino A, et al. Bronchiolar columnar cell dysplasia-genetic analysis of a novel preneoplastic lesion of peripheral lung. Virchows Arch 2003;442:429—36. Copin MC, Buisine MP, Devisme L, Leroy X, Escande F, Gosselin B, et al. Normal respiratory mucosa, precursor lesions and lung carcinomas: differential expression of human mucin genes. Front Biosci 2001;6:D1264—75. Chizhikov V, Chikina S, Gasparian A, Zborovskaya I, Steshina E, Ungiadze G, et al. Molecular follow-up of preneoplastic lesions in bronchial epithelium of former Chernobyl cleanup workers. Oncogene 2002;21:2398—405. Wistuba II, Behrens C, Milchgrub S, Bryant D, Hung J, Minna JD, et al. Sequential molecular abnormalities are involved in the multistage development of squamous cell lung carcinoma. Oncogene 1999;18:643—50. Thiberville L, Payne P, Vielkinds J, LeRiche J, Horsman D, Nouvet G, et al. Evidence of cumulative gene losses with progression of premalignant epithelial lesions to carcinoma of the bronchus. Cancer Res 1995;55:5133—9. Kayser K, Dunnwald D, Kazmierczak B, Bullerdiek J, Kaltner H, Zick Y, et al. Chromosomal aberrations, profiles of expression of growth-related markers including galectins and environmental hazards in relation to incidence of pulmonary hamartomas. Pathol Res Pract 2003;199:589—98. Kazmierczak B, Rosigkeit J, Wanschura S, Meyer-Bolte K, Van de Ven WJ, Kayser K, et al. HMGI-C rearrangements as the molecular basis for the majority of pulmonary chondroid hamartomas: a survey of 30 tumors. Oncogene 1996;12:515—21. Kayser K, Kayser G. TNM staging and survival of bronchus carcinoma patients. Elec J Pathol Histol 1998;4:906—82. Kayser K, Kayser G, Eichhorn S, Biechele U, Altiner M, Kaltner H, et al. Association of prognosis in surgically treated lung cancer patients with cytometric, histometric and lig- 214 [36] [37] [38] [39] K. Kayser et al. and histochemical properties: with an emphasis on structural entropy. Anal Quant Cytol Histol 1998;20:313—20. Kayser K, Bovin NV, Korchagina EY, Zeilinger C, Zeng FY, Gabius HJ. Correlation of expression of binding sites for synthetic blood group A-, B- and H-trisaccharides and for sarcolectin with survival of patients with bronchial carcinoma. Eur J Cancer 1994;30A:653—7. Kayser K, Becker C, Seeberg N, Gabius HJ. Quantitation of asbestos and asbestos-like fibers in human lung tissue by hot and wet ashing, and the significance of their presence for survival of lung carcinoma and mesothelioma patients. Lung Cancer 1999;24:89—98. Kayser K, Seemann C, Andre S, Kugler C, Becker C, Dong X, et al. Association of concentration of asbestos and asbestoslike fibers with the patient’s survival and the binding capacity of lung parenchyma to galectin-1 and natural alphagalactoside- and alpha-mannoside-binding immunoglobulin G subfractions from human serum. Pathol Res Pract 2000;196:81—7. Rubio CA. Two types of cells in the normal and atypical squamous epithelium of the cervix. II. Light microscopic study in human subjects. Acta Cytol 1982;26:121—5. [40] Hietanen S, Auvinen E, Grenman S, Lakkala T, Sajantila A, Klemi P, et al. Isolation of two keratinocyte cell lines derived from HPV-positive dysplastic vaginal lesions. Int J Cancer 1992;52:391—8. [41] Boyle JO, Lonardo F, Chang JH, Klimstra D, Rusch V, Dmitrovsky E. Multiple high-grade bronchial dysplasia and squamous cell carcinoma: concordant and discordant mutations. Clin Cancer Res 2001;7:259—66. [42] Kleeberger S. Genetic aspects of susceptibility to air pollution. Eur Respir J Suppl 2003;40:52s—6s. [43] Kayser G, Baumhakel JD, Szoke T, Trojan I, Riede U, Werner M, Kayser K. Vascular diffusion density and survival of patients with primary lung carcinomas. Virchows Arch 2003. [44] Miller RR. Alveolar atypical hyperplasia in association with primary pulmonary adenocarcinoma: a clinicopathological study of 10 cases. Thorax 1993;48:679—80. [45] Sterner DJ, Mori M, Roggli VL, Fraire AE. Prevalence of pulmonary atypical alveolar cell hyperplasia in an autopsy population: a study of 100 cases. Mod Pathol 1997;10:469—73. [46] Ritter JH. Pulmonary atypical adenomatous hyperplasia. A histologic lesion in search of usable criteria and clinical significance. Am J Clin Pathol 1999;111:587—9.