Loss of Chromosome 13q is Associated with Malignant Potential in ...

CANCER GENOMICS & PROTEOMICS 3: 39-46 (2006)

Loss of Chromosome 13q is Associated with Malignant Potential in Pulmonary Carcinoids KENTARO INAMURA1, REIKO FURUTA1, YUKITOSHI SATOH1,2, TAKASHI SHIRAKAWA1,5, SAKAE OKUMURA2, KEN NAKAGAWA2, MUTSUNORI FUJIWARA3, EIJU TSUCHIYA4 and YUICHI ISHIKAWA1 1Department

of Pathology, the Cancer Institute and 2Department of Chest Surgery, the Cancer Institute Hospital, Japanese Foundation for Cancer Research (JFCR), 3-10-6 Ariake, Koto-ku, Tokyo 135-8550; 3Department of Pathology, Japanese Red Cross Medical Center, 4-1-22 Hiroo, Shibuya-ku, Tokyo, 150-8935; 4Kanagawa Cancer Center Research Institute, 1-1-2 Nakao, Asahi-ku, Yokohama 241-0815; 5Current affiliation: Fujirebio Inc., 2-62-5, Nihombashi-hamacho, Chuo-ku, Tokyo, 103-0007, Japan

Abstract.

Background: Genetic differences between pulmonary typical carcinoids (TCs) and atypical carcinoids (ATCs) still remain controversial and no genetic marker is available for the evaluation of the malignant potential in pulmonary carcinoids. Materials and Methods: Five TCs and 3 ATCs were analyzed concurrently by comparative genomic hybridization (CGH) and cytogenetics to investigate the chromosomal abnormalities with reference to malignant potential. As parameters for biological aggressiveness, mitotic counts, necrosis, lymph and blood vessel invasion, lymph node and organ metastasis and Ki-67 index were used. Results: Using CGH analysis, chromosome 13q loss was observed in 4 cases, of which 3 were ATC cases and 1 was a TC case with aggressive features. All the cases without 13q loss showed no aggressive features. Upon immunohistochemical analysis, RB protein expression was detected regardless of the 13q status. Conclusion: The results of this study indicate that another tumor suppressor gene on 13q may be involved in the malignant potential in pulmonary carcinoids. Pulmonary carcinoids account for 25% of all carcinoids and about 1% of all lung tumors. The World Health Organization (WHO) divides them into typical carcinoids (TCs) and atypical carcinoids (ATCs), based on their mitotic counts (0-1/2 mm2 vs. 2-10/2 mm2, respectively) and

Correspondence to: Yuichi Ishikawa, Department of Pathology, ∆he Cancer Institute, Japanese Foundation for Cancer Research (JFCR), 3-10-6 Ariake, Koto-ku, Tokyo 135-8550, Japan. Tel: +813-35200111, Fax: +81-3-35700558, e-mail: [email protected] Key Words: Lung, carcinoid, chromosome 13q, comparative genomic hybridization, retinoblastoma protein.

1109-6535/2006 $2.00+.40

the absence or presence of necrosis (1). Currently, pulmonary carcinoids are considered to be part of a spectrum of pulmonary neuroendocrine tumors, ranging from low-grade malignant TC, to intermediate ATC, to high-grade large cell neuroendocrine carcinoma and small cell lung carcinoma. TCs rarely metastasize and show a 5year survival rate ranging from 87 to 99%, whereas patients with ATC have a greater tendency for metastasis and a significantly lower 5-year survival rate, ranging from 56 to 70% (2, 3). Little is known about the factors underlying the prognosis for pulmonary carcinoids, with the histological subdivision into TC and ATC, based on the morphological features described above, forming the current basis for assessment of prognosis (3). More recently, lymph node metastasis (2) and high expression of Ki-67 (4) have been reported to correlate with poor prognosis. Several groups have tried to genetically distinguish between TC and ATC using cytogenetic and molecular genetic methods or comparative genomic hybridization (CGH). Onuki et al. reported that loss of heterozygosity (LOH) at 5q21 correlated with poor survival in pulmonary carcinoids, showing a difference in LOH frequency between TCs (0%) and ATCs (25%) (5). Walch et al. used conventional CGH analysis to show that TCs and ATCs are both characterized by under-representation of 11q (TC, 48%; ATC, 67%), and also by further losses on 10q and 13q in ATCs, allowing cytogenetic differentiation between TCs and ATCs (6). More recently, Petzmann et al. and Ullmann et al. performed array CGH analysis and revealed that deletions of 11q are the most frequent chromosomal aberrations in ATCs but are rarely detected in TCs. These results were consistent with the data of their previous conventional CGH study (7, 8). The discrepancies between

39

CANCER GENOMICS & PROTEOMICS 3: 39-46 (2006) these studies have led to controversies concerning the differentiation of TCs from ATCs and the genetic difference between TCs and ATCs remains unclear. Consequently, no genetic marker is available for the evaluation of the malignant potential of a pulmonary carcinoid. The primary purpose of our study was to investigate the chromosomal abnormalities of pulmonary carcinoids with respect to malignant potential. We analyzed 5 TCs and 3 ATCs concurrently by conventional CGH and cytogenetics and looked for any chromosomal abnormalities that correlated with biological aggressiveness and expected to differ between TCs and ATCs. Biological aggressiveness was defined not only by mitotic counts and necrosis, but also lymph and blood vessel invasion, lymph node and organ metastasis and Ki-67 index. Our CGH results suggested that chromosome 13q loss is associated with the malignant potential in pulmonary carcinoids. Chromosome 13q includes the locus of the RB gene, an important tumor suppressor gene. To evaluate the involvement of the RB gene as the specific target of 13q loss, we performed immunohistochemical analysis of RB protein (pRB) expression.

Materials and Methods Five TC and 3 ATC cases were analyzed in this study. The histopathological classification was based on the WHO classification criteria (1). All clinical samples were collected with ethical committee approval and informed consent from patients undergoing surgery or polypectomy at the Cancer Institute Hospital, Tokyo, Japan, between June 1995 and March 1999. Their clinical profiles are summarized in Table I. Lobectomy was performed in 7 out of the 8 cases, whereas in 1 case (Case 3) the tumor located at the left upper bronchus was removed by bronchoscopic polypectomy. Although lymph node metastases were found in 3 cases, all the patients were alive with no recurrence for more than 5 years. For examinations by light microscopy, materials from the surgical or polypectomy specimens were fixed in 15% buffered formalin and embedded in paraffin. Sections, 4 Ìm thick, were stained with hematoxylin and eosin, as well as immunohistochemical markers including chromogranin-A, synaptophysin and N-CAM. As parameters for biological aggressiveness, mitotic counts, necrosis, lymph and blood vessel invasion, lymph node and organ metastasis and the Ki-67 index were used. Ki-67 expression, which correlates with tumor proliferative activity (9), was measured by immunohistochemical examination. The Ki-67 antibody detects a nuclear matrix-associated antigen that is present only in proliferating cells. In each case, a formalin-fixed, paraffin-embedded section was stained using the avidin-biotin peroxidase complex technique, with the monoclonal antibody against Ki-67 (clone MIB-1, DAKO, Glostrup, Denmark, 1:50). The Ki-67 index was defined as the percentage of tumor nuclei showing Ki-67 staining from a total of 1,000 neoplastic cells in areas with highest proliferative activity. CGH analysis. For the CGH analysis, the samples were grossly dissected and snap-frozen in liquid nitrogen within 20 min of removal

40

and stored at –80ÆC until analysis. In all these cases, sufficient normal lung or bronchial tissues from the same patients were available. Genomic DNA was extracted from the tumors and corresponding normal lung or bronchial tissues using the SepaGene Kit R (Sanko Junyaku Co. Ltd., Tokyo, Japan). Isolated tumor DNAs were labeled directly with Spectrum green-dUTP (Vysis, Downers Grove, IL, USA), and normal DNAs were labeled with Spectrum red-dUTP (Vysis) using nick translation. Labeled tumor and normal DNAs (200 ng each), together with 10 Ìg of normal Cot-1 DNA (Vysis), were denatured at 73ÆC for 5 min in 10 Ìl of hybridization solution (50% formamide, 10% dextran sulfate, 2 x SSC) and applied to normal lymphocyte metaphase spreads. The hybridization was performed according to the manufacturer’s instructions (Vysis) for 72 h at 37ÆC in a humid environment. Following hybridization, cover slips were removed, and the slides were washed 3 times for 10 min each in 2 x SSC/50% formamide. Slides were air-dried in the dark and counterstained with 10 Ìl of DAPI II (Vysis). Finally, the slides were sealed with nail polish and stored in the dark at 4ÆC until image acquisition. Three-color fluorescent images (4’,6-diamidino-2phenylindole, Spectrum green and red fluorescence) were collected from each metaphase spread using an epifluorescence microscope (Zeiss, Germany) coupled with a cooled charge-coupled device camera (Photometrics, Tucson, AZ, USA). The relative changes in the copy number of the DNA sequences were analyzed using a digital image analysis system (Quips-XL software; Vysis). The average ratios were calculated after automatically scaling the profiles of individual homologous chromosomes of the same length. Deviations from the average profiles were considered diagnostic for under- or overrepresentations of genomic DNA when the ratios were below 0.8 or above 1.2, respectively. Centromeric and telomeric regions, as well as sex chromosomes, were excluded from the analysis. Immunohistochemical analysis of pRB. Immunohistochemical detection of pRB was performed on formalin-fixed, paraffinembedded sections using the monoclonal antibody, clone G3-245 (BD PharMingen, San Diego, CA, USA, 1:50) recognizing an epitope between amino acids 332-344 of the human pRB. The 4-Ìm-thick sections were pretreated with 10 mM citrate buffer and by water bath immersion at 97ÆC. Incubation with the primary antibody was performed in a DAKO Auto-Stainer, with antibody localization visualized using a two-step method, the DAKO EnVision + System, horseradish peroxidase, and 3,3’-diaminobenzidine as a chromogen. Only nuclear positivity was assessed. Tumors were scored as pRBpositive if all tumor cells were stained or the staining pattern was heterogeneous with a portion of the tumor cells showing nuclear pRB staining. Tumors were scored pRB-negative if no tumor cells with nuclear staining were identified while adjacent normal stromal cells had nuclear pRB staining. Karyotyping. In 6 out of the 8 cases, karyotypes were examined with fresh materials using the Q-banding method. Without collagenase dissociation, a cell suspension was seeded into tissue culture containing RPMI-1640 medium and antibiotics, supplemented with 10% fetal bovine serum, and incubated at 37ÆC in a 5% CO2 atmosphere. When the cultures had entered exponential growth, they were harvested after Colcemid (0.01 to 0.05 Ìg/ml) treatment. Slides were prepared using conventional techniques and, after airdrying, underwent quinaklin mustard staining. The standard International System of Human Cytogenetic Nomenclature (ISCN, 1995) was used to identify chromosomal abnormalities (10).

Inamura et al: Chromosome 13q Loss in Pulmonary Carcinoids

Table I. Clinical profiles of the cases examined. Case ¡o. 1 2 3 4 5 6 7 8

Diagnosis

Age (years)

Sex

SIa

Locationb

Size (mm)

Stage

Follow-upc (years)

TC TC TC TC TC ATC ATC ATC

42 40 75 45 16 50 62 67

M M M M F M M M

0 480 NRd 440 0 20 1140 1410

mid-zonal mid-zonal central mid-zonal mid-zonal mid-zonal mid-zonal peripheral

16 48 NR 20 90 35 25 16

pT1N0M0 pT2N0M0 cT1N0M0 pT1N0M0 pT2N2M0 pT2N2M0 pT1N0M0 pT1N2M0

Alive (9.5) Alive (7) Alive (6) Alive (7) Alive (6) Alive (8) Alive (5) Alive (6.5)

aSI,

smoking index (defined as a product of the number of cigarettes per day and the duration (years)). of tumors: based on generation of bronchial branching. Central: from the first to the third bronchi (i.e. from main to subsegmental bronchi); mid-zonal: from the fourth to the sixth; peripheral: at the seventh bronchus or further. cAll the patients were alive with no evidence of recurrence. dNR, not recorded. TC, typical carcinoid ATC, atypical carcinoid bLocation

Results CGH analysis. A schematic summary of chromosomal gains and losses is shown in Figure 1. Chromosomal abnormalites were detected in 7 out of the 8 cases (87.5%) of pulmonary carcinoids. Only in 1 out of 5 of the TC cases were no abnormalities detected. In total, chromosomal losses (18 areas) were slightly more frequent than gains (15 areas). The most common abnormality was loss of chromosome 13q (4 out of 8 cases). Other areas with less frequent abnormalities included 1p, -3p, -11q, +7q, +16p, +17q (2 out of 8 cases each). Correlation of chromosomal abnormalities and biological aggressiveness. The parameters for biological aggressiveness with the summary of the chromosomal abnormalities detected by CGH are given in Table II. As expected, the Ki-67 indices were higher in the ATC (range, 7.5 to 16.8%) than in the TC cases (range, 0.2 to 0.9%), correlating with the more proliferative propensity of ATCs. One TC (Case 5) was deemed to be more aggressive than the other TCs according to the examined parameters. In this case, lymph node metastasis and microscopical vascular invasions (lymph and blood vessels) were found, whereas neither of them was observed in the other TC cases. The average number of abnormalities increased from 1.4 in the TC cases (range, 0 to 3) to 8.7 in ATC cases (range, 8 to 9). A 13q loss was observed in 4 cases, including all 3 ATC cases and 1 aggressive TC case. Interestingly, all the aggressive cases showed a 13q loss, while cases without aggressive features did not. Additional losses of 3p and 11q were observed only in the ATC (2 out of 3 cases each).

Immunohistochemical analysis of pRB. To evaluate the involvement of the RB gene as a specific target of chromosome 13q loss, pRB expression was examined by immunohistochemical staining. Out of the 8 cases, 1 (Case 3) was inadequate for pRB immunostaining, because normal stromal cells were not stained with pRB. All the remaining 7 cases, including those with chromosome 13q loss, showed positive pRB staining. All the cases demonstrated a mosaic staining pattern with positive and negative nuclei. Two ATC cases (Cases 6 and 8) showed stronger immunoreactivity than the other cases. In these 2 cases, the tumors showed many strongly-stained nuclei per high-power field, whereas the other cases showed mostly weak staining with only a few strongly-stained nuclei per high-power field (Figure 2). Karyotyping. Among the 6 cases examined for Q-banding analysis, no mitotic cells were detected in 2 TCs (Cases 3 and 4), normal karyotypes were demonstrated in 2 other TCs (Cases 1 and 5) and karyotypic abnormalities were observed in 2 ATCs (Cases 7 and 8) (Table II). In Case 7, the tumor cells showed monoclonal abnormalities in all 12 metaphases examined, resulting in the following karyotype: 46,XY, i(7)(q10),-8,del(10)(q24),-11,-13,i(17)(q10),+20, +mar1,mar2 (Figure 3a). Losses of chromosomes 11 and 13 were identical to the CGH data. Chromosome 7p loss and 7q gain detected by CGH were attributable to isochromosome 7q. Marker chromosomes may explain some other abnormalities identified by CGH. In Case 8, the following karyotype was obtained: 46,XY,t(12;15)(q13;q26) (Figure 3b). Only a single translocation was detected cytogenetically, although CGH identified more chromosomal abnormalites.

41

CANCER GENOMICS & PROTEOMICS 3: 39-46 (2006)

Figure 1. Schematic summary of chromosomal gains and losses detected in 5 typical carcinoids (blue lines) and 3 atypical carcinoids (red lines). Losses of tumor DNA are drawn on the left, gains on the right side.

It is noteworthy that tumor cells proliferated in vitro in both the ATC cases, whereas tumor cells showed no proliferation in half of the TC cases, in agreement with the proliferation activities measured by Ki-67.

Discussion The genetic difference between TCs and ATCs still remains controversial. No genetic marker is available for the evaluation of the malignant potential of a pulmonary carcinoid. In this study, we analyzed 5 TCs and 3 ATCs concurrently by conventional CGH technique and cytogenetics in order to investigate chromosomal abnormalities with regard to malignant potential. The CGH results suggest chromosome 13q loss as a candidate marker for malignant potential in pulmonary carcinoids. In 1 of the cases with 13q loss, chromosome 13 monosomy was identified by karyotyping. The cases with 13q loss comprised all the ATCs and 1 TC. The TC case

42

Figure 2. Immunohistochemical staining of pRB (magnification, x400). In all pulmonary carcinoids examined, including those with 13q loss, pRB immunostaining was observed. (a) In all typical carcinoid cases, weak immunoreactivity for pRB was detected. (b) In 2 out of the 3 atypical carcinoid cases, stronger immunoreactivity for pRB was detected.

with 13q loss turned out to be more aggressive than the other TC cases for the examined parameters. All the cases with no 13q loss showed no aggressive features. Several CGH studies have described the frequency of 13q loss in pulmonary carcinoids (Table IIIa), but the data are somewhat conflicting. This may be accounted for by the low reproducibility of ATCs (12). The histological separation of ATC from TC is sometimes difficult. In the study by Walch et al. (6), 13q loss occurred more frequently in the ATC cases (3 out of 6) than in the TC cases (3 out of 17). Furthermore, 1 of the 3 TC cases with 13q loss showed lymph node metastasis. Their results suggest a correlation between 13q loss and aggressive behavior of pulmonary carcinoids, in accordance with our results. Compared with other data, we found a higher frequency of 13q loss. This difference may be attributed to the small number of cases


Table II. Parameters for biological aggressiveness and a summary of chromosomal abnormalities detected by CGH and karyotyping. Case Dx Mitosis Necrosis ¡o. 2 mm2

ly

bl

Ki-67 Lymph Organ index node metastasis (%) metastasis

DNA losses on chromosomes

DNA gains on chromosomes

Karyotyping

1 2 3 4 5 6 7 8

– – – – + + – –

– – – – + – – +

0.7 0.3 0.3 0.9 0.2 7.5 11 16.8

* 1p * * 13q, 16q 1p, 3p, 3q, 9p, 9q, 11q, 13q 7p, 8p, 11q, 13q 3p, 4q, 13q, 18q

9q * 17q, 19p, 19q * * 11p, 16p 7q, 8q, 16p, 17q 1p, 5p, 7q, 8p, 13q

Normal karyotype Not examined No proliferation No proliferation Normal karyotype Not examined Abnormal karyotype Abnormal karyotype

TC TC TC TC TC ATC ATC ATC

0 0 0 0 0 1 4 9

– – – – – + + +

– – – – + + – +

– – – – – – – –

Dx, diagnosis; ly, lymph vessel invasion; bl, blood vessel invasion; TC, typical carcinoid; ATC, atypical carcinoid. *No detectable DNA changes.

in the present study, as well as the lack of histological reproducibility of ATCs. Chromosome 13q includes the locus of the well known RB tumor suppressor gene, loss of which is a key event in the initiation or progression of several human malignancies, including small cell lung carcinoma (13). LOH at the RB locus has previously been investigated in pulmonary carcinoids. In the study by Kobayashi et al. (14), no losses at this site were detected in the TC cases (0 out of 13) whereas the ATC cases demonstrated intermediate frequencies of LOH (2 out of 4), although conflicting results have been published (Table IIIb). To evaluate the involvement of the RB gene as the specific target of 13q loss, we performed an immunohistochemical analysis of pRB expression. In all pulmonary carcinoids adequately examined, including those with 13q loss, the presence of pRB was revealed, suggesting that RB is not the target of 13q loss. Another tumor suppressor gene on 13q may be involved in malignant potential in pulmonary carcinoids. Our immunohistochemical findings on the expression of pRB in pulmonary carcinoids are in keeping with the report by Cagle et al. (16). These authors investigated the pRB expression in neuroendocrine lung tumors and observed retention of pRB expression in TCs and ATCs, in contrast to loss in small cell carcinomas and large cell neuroendocrine carcinomas. The TCs in their study showed a weakly heterogeneous immunostaining pattern with few strongly-stained cells per high-power field, whereas many strongly-stained nuclei per high-power field were characteristic of these ATCs. In our study, we observed stronger immunoreactivity in 2 out of the 3 ATCs than in the TCs, a finding similar to theirs. In this study, pRB expression was observed in pulmonary carcinoids with 13q loss. A similar discrepancy between 13q abnormalities and pRB expression in pulmonary carcinoid cell lines has been reported (17). Gouyer et al. showed a lack of correlation between LOH at the RB locus and pRB

Figure 3. Q-banded karyotypes of two atypical carcinoid cases. (a) Case 7: 46,XY,i(7)(q10),-8,del(10)(q24),-11,-13,i(17)(q10),+20,+mar1,mar2 [12]. (b) Case 8: 46,XY,t(12;15)(q13;q26) [10].

43

CANCER GENOMICS & PROTEOMICS 3: 39-46 (2006) Table III. Summary of genetic data on typical carcinoids (TCs) and atypical carcinoids (ATCs) with special emphasis on chromosome 13q loss detected by CGH (a) and LOH at the RB locus (b). (a) Author

Number Percentage of cases (chromosome 13q loss)

Walch et al., 1998 (6) Zhao et al., 2000 (11) Ullmann et al., 2002 (8) This study

17 11 15 5

18% 0% 7% 20%

ATC Walch et al., 1998 (6) Ullmann et al., 2002 (8) This study

6 20 3

50% 10% 100%

TC

CGH, comparative genomic hybridization; LOH, loss of heterozygosity.

(b) Author

Number of cases

Percentage (LOH at the RB locus)

Gouyer et al., 1994 (15) Onuki et al., 1999 (5) Kobayashi et al., 2004 (14)

6 10 13

33% 20% 0%

ATC Gouyer et al., 1994 (15) Onuki et al., 1999 (5) Kobayashi et al., 2004 (14)

4 11 4

75% 22% 50%

TC

expression in pulmonary carcinoids (15). In addition to pulmonary carcinoids, a lack of concordance between LOH at the RB locus and absence of RB protein has been reported for several different tumors including breast, head and neck, prostate and ovarian cancers (18-22). These reports lend support to our view that there may be another tumor suppressor gene on 13q that correlates with malignant potential in pulmonary carcinoids. In addition to 13q loss, losses of 3p and 11q were observed specifically in the ATC cases. Ullmann et al. also observed these losses more frequently in ATCs than in TCs and suggested them as candidate markers for distinguishing ATCs from TCs (8). Losses of 3p and 11q also should be considered as candidate markers for malignant potential in pulmonary carcinoids. In summary, our CGH study suggests that chromosome 13q loss is associated with the malignant potential in pulmonary carcinoids. By immunohistochemical analysis, pRB expression was detected regardless of 13q status. We suppose that another tumor suppressor gene on 13q may be involved in the malignant potential in pulmonary carcinoids.

44

Acknowledgements The authors appreciate the supervision for karyotyping by Prof. Emeritus A. Tonomura of Tokyo Medical and Dental University, Japan, and thank Mr. Hiroyuki Mukai for karyotyping and Ms. Kazuko Yokokawa, Mr. Motoyoshi Iwakoshi, Ms. Miyuki Kogure and Ms. Tomoyo Kakita for their technical assistance and Ms. Chisato Kakuta for her secretarial support. Parts of this study were supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, and by grants from the Ministry of Health, Labour and Welfare, the Smoking Research Foundation, and the Vehicle Racing Commemorative Foundation, Japan.

References 1 Travis WD, Colby TV, Corrin B, Shimosato Y and Brambilla E: International Histological Classification of Tumours: Histological Typing of Lung and Pleural Tumours, 3rd ed. Berlin: Springer, 1999. 2 Cardillo G, Sera F, Di Martino M, Graziano P, Giunti R, Carbone L, Facciolo F and Martelli M: Bronchial carcinoid tumors: nodal status and long-term survival after resection. Ann Thorac Surg 77: 1781-1785, 2004. 3 Travis WD, Rush W, Flieder DB, Falk R, Fleming MV, Gal AA and Koss MN: Survival analysis of 200 pulmonary neuroendocrine tumors with clarification of criteria for atypical carcinoid and its separation from typical carcinoid. Am J Surg Pathol 22: 934-944, 1998. 4 Granberg D, Wilander E, Oberg K and Skogseid B: Prognostic markers in patients with typical bronchial carcinoid tumors. J Clin Endocrinol Metab 85: 3425-3430, 2000. 5 Onuki N, Wistuba II, Travis WD, Virmani AK, Yashima K, Brambilla E, Hasleton P and Gazdar AF: Genetic changes in the spectrum of neuroendocrine lung tumors. Cancer 85: 600-607, 1999. 6 Walch AK, Zitzelsberger HF, Aubele MM, Mattis AE, Bauchinger M, Candidus S, Prauer HW, Werner M and Hofler H: Typical and atypical carcinoid tumors of the lung are characterized by 11q deletions as detected by comparative genomic hybridization. Am J Pathol 153: 1089-1098, 1998. 7 Petzmann S, Ullmann R, Halbwedl I and Popper HH: Analysis of chromosome-11 aberrations in pulmonary and gastrointestinal carcinoids: an array comparative genomic hybridization-based study. Virchows Arch 445: 151-159, 2004. 8 Ullmann R, Petzmann S, Klemen H, Fraire AE, Hasleton P and Popper HH: The position of pulmonary carcinoids within the spectrum of neuroendocrine tumors of the lung and other tissues. Genes Chromosomes Cancer 34: 78-85, 2002. 9 Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U and Stein H: Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133: 1710-1715, 1984. 10 Mitelman F: Guidelines for Cancer Cytogenetics. Supplement to an International System for Human Cytogenetic Nomenclature. Basel: Karger, 1995. 11 Zhao J, de Krijger RR, Meier D, Speel EJ, Saremaslani P, Muletta-Feurer S, Matter C, Roth J, Heitz PU and Komminoth P: Genomic alterations in well-differentiated gastrointestinal and bronchial neuroendocrine tumors (carcinoids): marked differences indicating diversity in molecular pathogenesis. Am J Pathol 157: 1431-1438, 2000.


12 Travis WD, Gal AA, Colby TV, Klimstra DS, Falk R and Koss MN: Reproducibility of neuroendocrine lung tumor classification. Hum Pathol 29: 272-279, 1998. 13 Benedict WF, Xu HJ, Hu SX and Takahashi R: Role of the retinoblastoma gene in the initiation and progression of human cancer. J Clin Invest 85: 988-993, 1990. 14 Kobayashi Y, Tokuchi Y, Hashimoto T, Hayashi M, Nishimura H, Ishikawa Y, Nakagawa K, Sato Y, Takahashi A and Tsuchiya E: Molecular markers for reinforcement of histological subclassification of neuroendocrine lung tumors. Cancer Sci 95: 334-341, 2004. 15 Gouyer V, Gazzeri S, Brambilla E, Bolon I, Moro D, Perron P, Benabid AL and Brambilla C: Loss of heterozygosity at the RB locus correlates with loss of RB protein in primary malignant neuro-endocrine lung carcinomas. Int J Cancer 58: 818-824, 1994. 16 Cagle PT, el-Naggar AK, Xu HJ, Hu SX and Benedict WF: Differential retinoblastoma protein expression in neuroendocrine tumors of the lung. Potential diagnostic implications. Am J Pathol 150: 393-400, 1997. 17 Lai SL, Brauch H, Knutsen T, Johnson BE, Nau MM, Mitsudomi T, Tsai CM, Whang-Peng J, Zbar B, Kaye FJ and Gazdar AF: Molecular genetic characterization of neuroendocrine lung cancer cell lines. Anticancer Res 15: 225-232, 1995. 18 Borg A, Zhang QX, Alm P, Olsson H and Sellberg G: The retinoblastoma gene in breast cancer: allele loss is not correlated with loss of gene protein expression. Cancer Res 52: 2991-2994, 1992. 19 Cooney KA, Wetzel JC, Merajver SD, Macoska JA, Singleton TP and Wojno KJ: Distinct regions of allelic loss on 13q in prostate cancer. Cancer Res 56: 1142-1145, 1996.

20 Dodson MK, Cliby WA, Xu HJ, DeLacey KA, Hu SX, Keeney GL, Li J, Podratz KC, Jenkins RB and Benedict WF: Evidence of functional RB protein in epithelial ovarian carcinomas despite loss of heterozygosity at the RB locus. Cancer Res 54: 610-613, 1994. 21 Kim TM, Benedict WF, Xu HJ, Hu SX, Gosewehr J, Velicescu M, Yin E, Zheng J, D'Ablaing G and Dubeau L: Loss of heterozygosity on chromosome 13 is common only in the biologically more aggressive subtypes of ovarian epithelial tumors and is associated with normal retinoblastoma gene expression. Cancer Res 54: 605-609, 1994. 22 Yoo GH, Xu HJ, Brennan JA, Westra W, Hruban RH, Koch W, Benedict WF and Sidransky D: Infrequent inactivation of the retinoblastoma gene despite frequent loss of chromosome 13q in head and neck squamous cell carcinoma. Cancer Res 54: 4603-4606, 1994.

Received September 11, 2005 Accepted November 14, 2005

45