Large cell neuroendocrine carcinoma of the lung: a comparison with large cell carcinoma with neuroendocrine morphology and small cell carcinoma

Lung Cancer (2005) 47, 225—233

Large cell neuroendocrine carcinoma of the lung: a comparison with large cell carcinoma with neuroendocrine morphology and small cell carcinoma Wei-Xia Penga,∗, Takaaki Sanoa, Tetsunari Oyamaa, Osamu Kawashimab, Takashi Nakajimaa a

Department of Tumor Pathology, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan b Department of Chest Surgery, National Nishi-Gunma Hospital, 2854 Kanai, Shibukawa, Gunma 377-8511, Japan Received in revised form 21 June 2004 ; accepted 22 June 2004 KEYWORDS Large cell neuroendocrine carcinoma; Neuroendocrine morphology; p16; RB; 14-3-3 ␴; 34 ␤ E12; LOH at 3p

Summary Large cell neuroendocrine carcinoma (LCNEC) of the lung is a malignant neuroendocrine tumor clinicopathologically similar to and falling in-between atypical carcinoid tumor and small cell lung carcinoma (SCLC). The diagnosis of LCNEC is based mainly on a characteristic neuroendocrine morphology and biological neuroendocrine differentiation. In order to know the discrepancy between morphological and biological neuroendocrine differentiation, LCNEC was immunohistochemically and molecular biologically compared with large cell carcinoma with neuroendocrine morphology (LCCNM), which lacked only biological neuroendocrine differentiation among the criteria of LCNEC. Immunohistochemically, disruption of the RB pathway, namely a lack of RB expression and simultaneous overexpression of p16 protein, was characteristic of LCNEC, but not LCCNM. In G2/M cell cycle regulation, 14-3-3 ␴ expression was markedly reduced in LCNEC. Moreover, the antibody 34 ␤ E12 recognizing a set of large-sized keratin gave a different staining pattern between LCNEC and LCCNM. The immunohistochemical data suggested that LCNEC has a similar biological marker profile to SCLC and different from LCCNM. However, a loss of heterozygosity (LOH) analysis using microsatellite markers showed a high frequency of

* Corresponding author. Tel.: +81 27 220 7980; fax: +81 27 220 7981.

E-mail addresses: [email protected] (W.-X. Peng), [email protected] (T. Sano), [email protected] (T. Oyama), [email protected] (O. Kawashima), [email protected] (T. Nakajima)

0169-5002/$ — see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2004.06.018

226

W.-X. Peng et al. LOH at 3p in both LCNEC and LCCNM as well as in SCLC. Morphological neuroendocrine differentiation might not be identical to biological neuroendocrine differentiation in large cell carcinoma of the lung. © 2004 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Large cell neuroendocrine carcinoma (LCNEC) of the lung is a newly established clinicopathologic entity that is morphologically and biologically between atypical carcinoid and small cell carcinoma (SCLC) [1]. In 1999, precise histological criteria for LCNEC were proposed by the World Health Organization (WHO) based on the criteria proposed by Travis et al. [1]. With the WHO Histological Typing of Lung and Pleural Tumors, the clinicopathological features of this carcinoma have become evident [2]. That is, a large-scale histological review revealed that LCNEC accounted for 3.1% of tumors in 2790 patients undergoing resection for primary lung cancer and the overall 5-year survival was 57%, which is one of the poorest for subgroups of primary lung cancers [3]. The mean age of the patients was 65 and a male preponderance with a habitual smoking history was marked in clinical profiles of LCNEC [3—5]. LCNEC shares various clinicopathological and biological features with SCLC, but its identity is still controversial with respect to the effectiveness of chemotherapy [6—8]. According to the WHO histological classification of LCNEC, poorly differentiated carcinomas that do not fully satisfy the histological criteria of LCNEC can be subdivided into large cell carcinomas with neuroendocrine differentiation lacking a neuroendocrine morphology, large cell carcinomas with a neuroendocrine morphology lacking neuroendocrine differentiation (LCCNM), classic large cell carcinoma and non-SCLC with neuroendocrine differentiation [2]. In routine pathological diagnosis, however, it is somewhat difficult to differentiate these histological subtypes from true LCNEC, and the clinicopathological significance of these histological subtypes still remains unclear. When differentiating LCNEC from poorly differentiated carcinomas and large cell carcinomas, confirmation of neuroendocrine morphology and neuroendocrine differentiation is critical among the other histological criteria of LCNEC proposed by the WHO Histological Classification [2]. In this study, therefore, we first collected poorly differentiated carcinomas and large cell carcinomas showing neuroendocrine morphology, from which LCNEC was selected based on the immunohistochemistry using neuroendocrine markers. We obtained two

histological subtypes of poorly differentiated lung cancers: LCNEC and LCCNM. In this study, we aimed to clarify the biological difference between LCNEC and LCCNM in terms of immunohistochemistry and loss of heterozygosity (LOH) at 3p, and concluded that LCNEC had various similarities to SCLC, but differed from LCCNM in several biological aspects, especially cell cycle regulation for tumor growth.

2. Materials and methods 2.1. Case selection The pathological files of lung cancer patients undergone operation at the Gunma University Hospital and the National Sanatorium Nishi-Gunma Hospital between 1990 and 2003 were histologically reviewed. Of 160 surgically resected tumors originally diagnosed as LCNEC, large cell carcinoma and poorly differentiated squamous cell and adenocarcinomas, 25 were selected as candidates for LCNEC to be reclassified according to the WHO criteria, and used in this study. The age distribution was 57—78 (mean: 71) years and the sex ratio was 23:2.

2.2. Immunohistochemistry Sections 3 ␮m thick were cut from paraffin blocks of the tumor and placed on silane-coated glass slides for use in the immunohistochemical analysis. The sections were dewaxed in xylene and rehydrated in graded a series of ethanol solutions, and the endogenous peroxidase activity was blocked with a 0.3% H2 O2 —methanol solution. Prior to the primary antibody reaction, the slides were subjected to antigen retrieval as described in Table 1. The slides were rinsed in 0.01 mol/L of phosphatebuffered saline (PBS), pH 7.4, and the sections were incubated with 10% normal serum of various animal species adjusted to that of the secondary antibody for 30 min in order to reduce non-specific immunostaining. In order to confirm the biological nature of the candidate for LCNEC in this study, the primary antibodies listed in Table 1 were applied to the slides overnight at 4 ◦ C. The primary antibody for cyclin D1 was incubated overnight at room temperature. After being thor-

Large cell neuroendocrine carcinoma of the lung

Table 1

227

Primary antibodies used in this study

Name

Clone or code

Antigen retrevala

Dilution

Company

NSE NCAM (CD56) Synaptophysin Chromogranin A p53 p21 p16

BBS/NC/VI-H14 1B6 SY38 LK2H10 DO-7 4D10 JC8

None C C None A D B

×200 ×100 ×20 ×100 ×50 ×40 ×50

pRB Cyclin D1 14-3-3 ␴

G3-24S DCS-fl, Code no: 18842

B D None

×200 ×50 ×100

Keratin Vimentin 34 ␤ E12 TTF-1 SAP CEA C-kit

KL1 M725 34 ␤ E12 8G7Q3/1 M4501 COL-1 Code no: 18101

None None E D None None C

Ready to use Ready to use ×10 ×100 ×50 Ready to use ×200

FHIT

Code no: 18163

B

Ready to use

Dako, Gtestrup, Denmark Novocastra, Laboratories, Newcastle, UK Dako, Glosrtrup, Denmark Lipsnaw, Michtoan, USA Novocastra Laboratories, Newcastle, UK Movocastra Laboratories, Newcastle, UK Kindly provided by Dr. J. Koh, Massachusetts General Hospital Cancer Center, Boston, MA, USA Pharmmgen, San Diego, CA. USA Movocastra Laboratories, Newcastle, UK Immuno-Biological Laboratory, Fujioka, Gunma, Japan Immunoteqh, Marseille, France Dako, Gktstrup, Denmark Dako, Gtestrup Denmark NeoMarkers, Fremont, CA, USA Dako, Glostrup, Denmark Dako, Glostrup, Denmark Immuno-Biological Laboratory, Fuaoka, Gunma, Japan Immuno-Biological Laboratory, Fuaoka, Gunma, Japan

a

A: antigen retrieval with 20% ZnSO4 solution for 20 min at 98 ◦ C, B: antigen retrieval with 0.01 M sodium phosphate-citric acid buffer, pH 8.0, for 20 min at 98 ◦ C, C: antigen retrieval with 0.1 M citric acid buffer, pH 6.0, for 20 min at 98 ◦ C, D: 0.001 M ethylenediaminetetraacetate (EDTA) solution, pH 8.0, for 20 min at 98 ◦ C, E: antigen retrieval with 0.05% pronase solution for 30 min at room temperature, F: antigen retrieval with 0.1% trypsin solution for 30 min at room temperature.

oughly washed with PBS containing 0.1% Triton X100, the slides were incubated with biotinylated secondary antibody for 30 min followed by a 1:100 dilution of the avidin—biotin—peroxidase complex (VECTASTAIN, Vector Laboratories, Burlingame, CA, USA) or streptavidin-HRP (for cyclin D1 and p21) for a further 30 min. The peroxidase was visualized with 0.02% 3-3 -diaminobenzidine tetrahydrochloride containing 0.005% H2 O2 in 0.01M Tris-phosphate buffer, pH 7.4. Finally, the sections were counterstained lightly with hematoxylin. Immunohistochemical results were evaluated as negative (no positive cells), mild (less than 5% of cells positive; ±), partial (5—50% of cells positive; +) and diffuse (more than 50% of cells positive; ++).

2.3. LOH analysis on chromosome 3 To extract DNA from tumors and normal tissues, several paraffin sections containing only tumor tissue or normal lung parenchyma were macroscopically dissected from the paraffin blocks after marking the area, and placed in an Eppendorf tube. Then, paraf-

fin sections were dewaxed with xylene, placed in an absolute ethanol solution and completely dried up in a speed back concentrator. For tissue digestion, a proteinase K solution containing 10% SDS was put into the Eppendorf tubes and kept at 60 ◦ C for 48 h with shaking. After the tissue was completely digested, DNA was purified by the standard phenol—chloroform extraction method. Finally, the DNA was adjusted to a concentration of 100 ␮g/ml with TE. To evaluate LOH on the 3p arm, three microsatellite polymorphic markers, D3S1234 (3p14.2), D3S4622 (3p21.3) and D3S1597 (3p25), were selected. Primer sequences were obtained from the Genome Database (http://www.gdb.org) for all these markers. PCR was performed in a 20 ␮l volume of a mixture containing 1.5 mM MgCl, 0.5 ␮M of each primer, 0.5 mM each of dATP, dGTP, and dTTP and 0.05 mM of dCTP, 0.2 ␮l of hot start Taq DNA polymerase (QIAGEN, Valencia, CA) and 0.1 ␮l of ␣32 P dCTP (3000 Ci/mmol, 10 mCi/ml). PCR was carried out for 40 cycles using a GeneAMP PCR System 9600 (Perkin-Elmer Cetus, Norwalk, CT, USA) after an initial denaturation for 15 min at 95 ◦ C. Each cycle consisted of denaturation for 30 s at 94 ◦ C, annealing for 30 s at 55 ◦ C (D3S1234 and D3S1597) or

228 57 ◦ C (D3S4622), and extension for 1 min at 72 ◦ C. The final extension was for 5 min at 72 ◦ C. The PCR products were denatured for 5 min at 98 ◦ C, electrophoresed on 5% denaturing polyacrylamide gels (3 mm thick) for 2 h at 1500 V, and visualized by exposure to X-ray films after drying. LOH was detected by comparing the normal allele pattern with the tumor one. As a control for the LOH analysis, five randomly selected small cell carcinomas were used.

2.4. Statistical analysis Fisher’s exact tests were used for statistical analyses of the results. P-values less than 0.05 were regarded as significant.

3. Results 3.1. Diagnosis of LCNEC and LCCNM Among 25 cases, 16 tumors were positive for at least one of three neuroendocrine markers, chromogranin A, synaptophysin and neural cell adhesion molecule (NCAM). These 16 cases were regarded as LCNEC, and the remaining nine cases were grouped as LCCNM. A precise histological analysis revealed that LCCNM had been previously diagnosed as large cell carcinoma (five cases), poorly differentiated squamous cell carcinoma (three) and adenocarcinoma (one). In LCNEC, tumors were positive for chromogranin A, synaptophysin and NCAM (Fig. 1A) in 4, 10 and 11 cases, respectively. Simultaneous synaptophysin and NCAM positivity was seen in five cases, in four of which tumors were also positive for chromogranin A. On the other hand, the general neuroendocrine marker neurone-specific enolase (NSE) was detected in 81.3 and 55.6% of LCNECs and LCCNMs, respectively.

3.2. Immunohistochemical comparison between LCNEC and LCCNM To know the difference in the biological nature of LCNEC and LCCNM, the expression of various proteins was examined immunohistochemically. The results are shown in Table 2. In keratin expression, the antibody 34 ␤ E12 produced negative results in most LCNECs, which contrasted with the positive results in LCCNM (Fig. 1B and C). The frequency of intermediate-sized keratin filament expression detected with the KL1 antibody and of vimentin expression revealed no difference between

W.-X. Peng et al. them. Immunohistochemically, each LCNEC and LCCNM showed almost or completely negative results for surfactant apoprotein (SAP) and fragile histidine triad (FHIT) protein. Of nine LCNECs positive for carcinoembryonic antigen (CEA), eight cases showed diffuse immunostaining (++) in the cytoplasm. Immunoreactivity for c-kit was present in the cell membrane and a diffuse staining pattern (++) was seen in half of both LCNEC and LCCNM (Fig. 1D). Also, diffuse nuclear immunostaining of thyroid transcription factor-1 (TTF-1) was observed in 31% of LCNECs and 22% of LCCNMs. As to the expression of cell cycle regulatory proteins (Table 3), strong and diffuse immunoreactivity for p16 was observed in both the nucleus and cytoplasm in LCNEC, which markedly differed from the results for LCCNM (Fig. 1E and F). In contrast with p16 expression, pRB immunoreactivity in the nucleus was less frequent in LCNEC than in LCCNM (Fig. 1G and H). P16(++)/pRB(−) and p16(−)/pRB(++) staining patterns were observed in 10 LCNECs and 6 LCCNMs, respectively. Subsequently, p16 and pRB expression was inversely correlated between LCNEC and LCCNM (P < 0.05). In the expression of p53 and p21, no difference was observed between LCNEC and LCCNM. However, a diffuse staining pattern for p53 (Fig. 1I) was observed in seven of nine LCNEC positive for p53 and no diffuse staining pattern was seen in LCCNM. Immunoreactivity for cyclin D1 was detected in the nucleus and there was no difference in cyclin D1 expression between LCNEC and LCCNM (Fig. 1J). However, all cyclin D1-positive tumors also expressed pRB in both LCNEC and LCCNM. 14-3-3 ␴ was detected immunohistochemically mainly in the cytoplasm. In LCNEC, 14-3-3 ␴ expression was markedly decreased as compared with that in LCCNM (Fig. 1K and L) (P < 0.05).

3.3. LOH on chromosome 3 The results of the LOH analysis of 3p are summarized in Fig. 2. In LCNEC and SCLC, the frequency of LOH at microsatellite markers D3S1597 (3p25), D3S4622 (3p21.3) and D3S1234 (3p14.2) was 63.6 and 75%, 63.6 and 66.7%, and 50 and 100%, respectively. In LCCNM, it was 50, 60 and 66.7%, respectively. Excluding non-informative cases with homozygosity, LOH on 3p involving one or more microsatellite markers was detected in 13 (81.3%) of 16 LCNECs, 6 (66.7%) of nine LCCNMs and 5 (100%) of five SCLCs. Although there were no significant differences in the presence of LOH between LCNEC and LCCNM, the frequency of LOH at 3p was highest in SCLC followed by LCNEC and then LCCNM.

Large cell neuroendocrine carcinoma of the lung

229

Fig. 1. Immunohistochemical results in large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology. LCNEC was diffusely positive for NCAM (A), but completely negative for 34 ␤ E12 (B). However, 34 ␤ E12 was frequently expressed in LCCNM (C). In more than half of LCNEC, c-kit was diffusely positive on the cell membrane of tumor cells (D). In cell cycle regulators, p16 was diffusely positive in LCNEC (E), but not in LCCNM (F). In LCNEC, pRB expression was lacked completely (G). Inversely, LCCNM showed marked pRB expression (H). Diffuse nuclear staining for p53 (I) and cyclin D1 (J) was observed in LCNEC. In LCNEC, 14-3-3 ␴ was completely negative (K), but LCCNM showed strong expression of 14-3-3 ␴ (L).

4. Discussion Even with the separation of LCNEC from large cell carcinoma, there still remain several large cell carcinomas morphologically and biologically similar to LCNEC. LCCNM must be distinguished from LCNEC from the standpoint of true neuroendocrine differ-

entiation. It remains unclear whether LCCNM is a new tumor subtype or different clinical entity from LCNEC. To our knowledge, only two papers describing LCCNM have appeared so far, and these clinicopathological studies concluded that LCNEC and LCCNM show similar clinical behavior and are more aggressive than ordinary LCC of the lung [4,9]. As no

230

Table 2 Group LCNEC n = 16 LCCNM n=9

W.-X. Peng et al.

Immunohistochemical results of various biological markers ∗

34 ␤ E12 2 (12.5) 6 (66.7)

KL1

VIM

14 (87.5) 7 (77.8)

1 (6.25) 2 (22.2)

TTF-1 5 (31.3) 2 (22.2)

SAP

CEA

c-kit

FHIT

1 (6.25) 0 (0)

9 (56.3) 5 (55.6)

10 (62.5) 4 (44.4)

0 (0) 0 (0)

The parenthesis means the percent of positive staining. ∗ P < 0.01.

Table 3

Immunohistochemical results of cell cycle regulator proteins

Histology

∗

LCNEC n = 16 LCCNM n=9

12 (75) 2 (22.2)

p16

∗∗

RB

6 (37.5) 8 (88.9)

p53

p21

9 (56.3) 6 (66.7)

9 (56.3) 5 (55.6)

cyclinD1 6 (37.5) 5 (55.6)

∗∗∗

14-3-3 ␴ 3 (18.8) 8 (88.9)

The parenthesis means the percent of positive staining. ∗ P < 0.05. ∗∗ P < 0.05. ∗∗∗ P < 0.01.

immunohistochemical or genetic studies have been carried out on both LCNEC and LCCNM, this is the first study to elucidate the biological difference between the two tumors. LCCNM was clinicopathologically quite similar to LCNEC. The age distribution is almost the same as that of non-SCLC and male preponderance is marked in LCNEC and LCCNM [3,4,9]. In this study, LCNECs were formally diagnosed as poorly differentiated squamous cell carcinomas and, in fact, previous reports pointed out that LCNEC was frequently misdiagnosed as poorly differentiated squamous cell carcinoma or poorly differentiated adenocarcinoma [3,10]. To distinguish LCNEC from LCCNM, we used the immunohistochemical detection of three neuroendocrine markers, chromogranin A, NCAM and synaptophysin, which are the most reliable neuroendocrine markers

in terms of sensitivity and specificity [11]. Chromogranin A has the highest specificity, because all chromogranin A-positive cases also showed positive staining for both NCAM and synaptophysin in this study. A previous study found that peptide hormone was a specific marker for neuroendocrine differentiation, but its sensitivity was low [10]. We clearly demonstrate that the growth inhibitory RB pathway in the regulation of the cell cycle is disrupted in LCNEC, but not LCCNM. The disruption of the RB pathway, which is primarily caused by a RB gene abnormality and results in a compensatory overexpression of p16 through the mediation of E2Fl, is a characteristic feature of SCLC as well as LCNEC [12—15]. It is well recognized that abnormality of the RB gene in SCLC is caused by either a deletion or mutation in both alleles. On the other hand, the frequent disruption of the RB pathway in

Fig. 2. The results of LOH analysis on chromosome 3p using microsatellite markers. The LOH at 3p was the most frequent in SCLC followed by LCNEC and then LCCNM.

Large cell neuroendocrine carcinoma of the lung non-SCLC is caused by an abnormality of the p16 gene itself rather than the RB gene [16—19]. The p16 gene was frequently inactivated in non-SCLC, especially in squamous cell carcinoma, mainly by a homozygous deletion, promoter methylation, or mutation, which resulted in negative immunohistochemical results [19,20]. In terms of the abnormality of the RB pathway, LCNEC is clearly a different tumor from LCCNM. In addition to abnormality of the RB pathway, 14-3-3 ␴, a cell cycle regulator of the G2/M checkpoint, showed a different pattern of immunohistochemical expression between LCNEC and LCCNM in this study. Previous studies have already revealed that 14-3-3 ␴ expression is markedly reduced in neuroendocrine lung cancers of both SCLC and LCNEC and causes de-regulation of the G2/M checkpoint [21,22]. Down —regulation of 14-3-3 ␴ expression is caused by epigenetic hypermethylation of the 14-3-3 ␴ promoter region and this phenomenon is seen in several human cancers, mainly in breast cancers and neuroendocrine carcinoma of the lung [21—24]. In lung cancers, our previous immunohistochemical study has revealed that 14-3-3 ␴ is not only expressed in adenocarcinoma, but also overexpressed in squamous cell carcinoma [25]. Similar to non-SCLC, colon and pancreatic cancers have recently been reported to express 14-3-3 ␴ protein without any hypermethylation of the promoter region [26]. Therefore, the expression pattern of 143-3 ␴ in LCCNM is similar to that in non-SCLC. The tumor suppressor p53 is an important cell cycle regulator affecting the G1 checkpoint through the p53 pathway and G2/M checkpoint through mainly 14-3-3 ␴ expression [27]. As shown in Table 3, no difference in p53 expression was observed between LCNEC and LCCNM. However, a diffuse staining pattern, suggesting a point mutation of the p53 gene, was seen in LCNEC. Disruption of the p53 pathway due to p53 abnormality is a characteristic feature of SCLC and squamous cell carcinoma, but not adenocarcinoma, especially the well differentiated type [28—31]. There was no significant difference in cyclin D1 expression between LCNEC and LCCNM in this study. However, cyclin D1 expression was decreased in highly malignant neuroendocrine carcinomas [13]. Among the immunohistochemical panel used to confirm the biological characteristics of the cell lineage in both LCNEC and LCCNM, a set of high molecular-weight cytokeratins (CK1, 5, 10 and 14) recognized by the antibody 34 ␤ E12 was characteristically absent in most LCNECs and its expression pattern significantly differed from that in LCCNM. In the lung, a complete lack of 34 ␤ E12 expression is characteristic of the neuroendocrine

231 cell lineage including neuroendocrine cell hyperplasia, tumorlets, typical and atypical carcinoid tumors, LCNEC and SCLC [32]. Therefore, the absence of 34 ␤ E12 expression and presence of TTF-1 expression is quite useful for the differentiation of LCNEC from morphologically similar lung cancers such as basaloid carcinoma and basaloid variant of squamous cell carcinoma [33]. TTF-1 is expressed in LCNEC and SCLC as well as in adenocarcinoma, but not in squamous cell carcinoma [34]. The expression in adenocarcinoma is reasonable, because TTF-1 activates the transcription of surfactant apoproteins and Clara cell protein [35,36]. However, TTF-1 expression in neuroendocrine tumors of the lung is controversial, because TTF-1 is not expressed in neuroendocrine cell hyperplasia, tumorlets or typical and atypical carcinoid tumors [34,37]. Recently, overexpression of c-kit protein or CD 117 has been reported in LCNEC and SCLC, and these c-kit-positive cancers are expected to be treatable with the new drug STI571, a c-kit tyrosine kinase inhibitor [38—41]. Of all the proteins of the immunohistochemical panel used in this study, however, only 34 ␤ E12 was an useful marker for distinguishing LCNEC from LCCNM. The WHO Histological Classification describes that LOH at 3p, 13q14, 9p and 5q22 has been found in all neuroendocrine tumor types, with an increasing frequency from typical carcinoid to LCNEC and SCLC, but with no difference between LCNEC and SCLC [2]. In lung cancers, tumor suppressor genes at 3p are suspected to play a critical role in the pathogenesis of both SCLC and non-SCLC [42]. On chromosome 3, there are several known genes such as FHIT, RASSF1A and VHL etc, but a true tumor suppressor gene involved in lung cancer development is poorly understood despite extensive studies [42—48]. The results of our LOH analysis of 3p corroborated previous reports in SCLC and LCNEC and also revealed a high frequency of LOH at 3p in LCCNM. Immunohistochemically, no FHIT protein was detected in LCNEC or LCCNM, which might reflect the results of the LOH analysis. Previous immunohistochemical studies of FHIT protein have shown a greatly reduced expression in SCLC and squamous cell carcinoma [49].

5. Conclusion Biologically, LCNEC was similar to SCLC, but differed from LCCNM, especially in cell cycle regulation, therefore morphological neuroendocrine differentiation in LCC might not be always identical to true neuroendocrine differentiation. Further

232

W.-X. Peng et al.

studies are necessary to establish the entity of LCCNM from LCC. [12]

Acknowledgements This work was supported in part by a Grant-inAid for Scientific Research from the Ministry of Education, Science, Sports and Culture, and by a Grant-in-Aid for Cancer Research from the Ministry of Health, Labor, and Welfare, Japan.We thank Drs. H Iijima and Y Tomizawa (Department of Medicine and Molecular Science, Gunma University Graduate School of Medicine) for technical advices and valuable suggestions, and also thank Mr. F. Hara, Ms. M. Saito and Mr. T. Hikino in our department for their technical assistance.

[13]

[14]

[15] [16]

[17]

References [1] Travis WD, Linnoila RI, Tsokos MG, Hitchcock CL, Cutler GB, Nieman L, et al. Neuroendocrine tumors of the lung with proposed criteria for large-cell neuroendocrine carcinoma. Am J Surg Pathol 1991;15:529—53. [2] World Health Organization. Histological typing of lung and pleural tumors. 3rd ed. Berlin: Spriner-Verlag; 1999. [3] Takei H, Asamura H, Maeshima A, Suzuki K, Kondo H, Niki T, et al. Large cell neuroendocrine carcinoma of the lung: a clinicopathologic study of 87 cases. J Thorac Cardiovasc Surg 2002;124:285—92. [4] Iyoda A, Hiroshima K, Toyozaki T, Haga Y, Fujisawa T, Ohwada H. Clinical characterization of pulmonary large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology. Cancer 2001;91:1992—2000. [5] Mazieres J, Daste G, Molinier L, Berjaud J, Dahan M, Delsol M, et al. Large cell neuroendocrine carcinoma of the lung: pathological study and clinical outcome of 18 resected cases. Lung Cancer 2002;37:287—92. [6] Przygodzki RM, Finkelstein SD, Langer JC, Swalsky PA, Fishback N, Bakker A, et al. Analysis of p53, K-ras-2, and C-raf-1 in pulmonary neuroendocrine tumors. Correlation with histological subtype and clinical outcome. Am J Pathol 1996;148:1531—41. [7] Onuki N, Wistuba II, Travis WD, Virmani AK, Yashima K, Brambilla E, et al. Genetic changes in the spectrum of neuroendocrine lung tumors. Cancer 1999;85:600—7. [8] Ullmann R, Schwendel A, Klemen H, Wolf G, Petersen I, Popper HH. Unbalanced chromosomal aberrations in neuroendocrine lung tumors as detected by comparative genomic hybridization. Hum Pathol 1998;29:1145—9. [9] Zacharias J, Nicholson AG, Ladas GP, Goldstraw P. Large cell neuroendocrine carcinoma and large cell carcinomas with neuroendocrine morphology of the lung prognosis after complete resection and systematic nodal dissection. Ann Thorac Surg 2003;75:348—52. [10] Jiang SX, Kameya T, Shoji M, Dobashi Y, Shinada J, Yoshimura H. Large cell neuroendocrine carcinoma of the lung: a histologic and immunohistochemical study of 22 cases. Am J Surg Pathol 1998;22:526—37. [11] Harada M, Yokose T, Yoshida J, Nishiwaki Y, Nagai K. Immunohistochemical neuroendocrine differentiation is

[18]

[19]

[20] [21]

[22]

[23] [24]

[25]

[26]

[27] [28]

an independent prognostic factor in surgically resected large cell carcinoma of the lung. Lung Cancer 2002;38: 177—84. Kelley MJ, Nakagawa K, Steinberg SM, Mulshine JL, Kamb A, Johnson BE. Differential inactivation of CDKN2 and Rb protein in non-small-cell and small-cell lung cancer cell lines. J Natl Cancer Inst 1995;87:756—61. Gugger M, Burckhardt E, Kappeler A, Hirsiger H, Laissue JA, Mazzucchelli L. Quantitative expansion of structural genomic alterations in the spectrum of neuroendocrine lung carcinomas. J Pathol 2002;196:408—15. Hirotoshi DA, Cagle PT, Hiroumi H, Fujita M, Yamashita M, Sharma A, et al. Differential retinoblastoma and p16INK4A protein expression in neuroendocrine tumors of the lung. Cancer 2000;88:550—6. Yuan J, Knorr J, Altmannsberger M, Goeckenjan G, Scharl A, Strebhardt K. Expression of p16 and lack of pRB in primary small cell lung cancer. J Pathol 1999;189:358—62. Kratzke RA, Greatens TM, Rubins JB, Maddaus MA, Niewoehner DE, Niehans GA, et al. Rb and p16INK4a expression in resected non-small cell lung tumors. Cancer Res 1996;56:3415—20. Sakaguchi M, Fujii Y, Hirabayashi H, Yoon HE, Komoto Y, Oue T, et al. Inversely correlated expression of p16 and Rb protein in non-small cell lung cancers: an immunohistochemical study. Int J Cancer 1996;65:442—5. Tanaka H, Fujii Y, Hirabayashi H, Miyoshi S, Sakaguchi M, Yoon HE, et al. Disruption of the RB pathway and cellproliferative activity in non-small-cell lung cancers. Int J Cancer 1998;79:111—5. Kashiwabara K, Oyama T, Sano T, Fukuda T, Nakajima T. Correlation between methylation status of the p16/CDKN2 gene and the expression of p16 and Rb proteins in primary non-small cell lung cancers. Int J Cancer 1998;79:215—20. Gazzeri S, Gouyer V, Vour’ch C, Brambilla C, Brambilla E. Mechanisms of p16INK4A inactivation in non small-cell lung cancers. Oncogene 1998;16:497—504. Osada H, Tatematsu Y, Yatabe Y, Nakagawa T, Konishi H, Harano T, et al. Frequent and histological type-specific inactivation of 14-3-3␴ in human lung cancers. Oncogene 2002;21:2418—24. Yatabe Y, Osada H, Tatematsu Y, Mitsudomi T, Takahasi T. Decreased expression of 14-3-3␴ in neuroendocrine tumors is independent of origin and malignant potential. Oncogene 2002;21:8310—9. Umbricht CB, Evron E, Gabrielson E, Ferguson A, Marks J, Sukumar S. Hypermethylation of 14-3-3␴ (stratifin) is an early event in breast cancer. Oncogene 2001;20:3348—53. Ferguson AT, Evron E, Umbricht CB, Pandita TK, Chan TA, Hermeking H, et al. High frequency of hypermethylation at the 14-3-3␴ locus leads to gene silencing in breast cancer. Proc Natl Acad Sci USA 2000;97:6049—54. Nakajima T, Shimooka H, Peng WX, Segawa A, Motegi A, Zhang J, et al. Immunohistochemical demonstration of 143-3␴ protein in normal human tissues and lung cancers, and the preponderance of its strong expression in epithelial cells of squamous cell lineage. Pathol Int 2003;53:353—60. Ide M, Nakajima T, Asao T, Kuwano H. Inactivation of 14-3-3␴ by hypermethylation is a rare event in colorectal cancers and its expression may correlate with cell cycle maintenance at the invasion front. Cancer Lett [in press]. Hermeking H, Lengauer C, Polyak K, He TC, Zhang L, Thiagalingam S, et al. 14-3-3␴ is a p53-regulated inhibition of G2/M progression. Mol Cell 1997;1:3—11. Takahashi T, Nau MM, Chiba I, Birrer MJ, Rosenberg RK, Vinocour M, et al. p53: a frequent target for genetic abnormalities in lung cancer. Science 1989;246:491—4.

Large cell neuroendocrine carcinoma of the lung [29] D’Amico D, Carbone D, Mitsudomi T, Nau M, Fedorko J, Russell E, et al. High frequency of somatically acquired p53 mutations in small-cell lung cancer cell lines and tumors. Oncogene 1992;7:339—46. [30] Fong KM, Kida Y, Zimmerman PV, Ikenaga M, Smith PJ. Loss of heterozygosity frequently affects chromosome 17q in non-small cell lung cancer. Cancer Res 1995;55:4268—72. [31] Xue Q, Sano T, Kashiwabara K, Oyama T, Nakajima T. Aberrant expression of pRb, p16, p14ARF, MDM2, p21 and p53 in squamous cell carcinomas of lung. Jpn J Cancer Res 2001;92:285—92. [32] Sturm N, Rossi G, Lantu´ ejoul S, Laverri` ere MH, Papotti M, Brichon PY, et al. 34E12 expression along the whole spectrum of neuroendocrine proliferations of the lung, from neuroendocrine cell hyperplasia to small cell carcinoma. Histopathology 2003;42:156—66. [33] Sturm N, Lantu´ ejoul S, Laverri` ere MH, Papotti M, Brichon PY, Brambilla C, et al. Thyroid transcription factor 1 and cytokeratins 1, 5, 10, 14 (34 ␤ E12) expression in basaloid and large-cell neuroendocrine carcinomas of the lung. Hum Pathol 2001;32:918—25. [34] Zamecnik J, Kodet R. Value of thyroid transcription factor-1 and surfacntant apoprotein A in the differential diagnosis of pulmonary carcinomas: a study of 109 cases. Virchows Arch 2002;440:353—61. [35] Zhang L, Whitsett JA, Stripp BR. Regulation of Clara cell secretory protein gene transcription by thyroid transcription factor-1. Biochim Biophys Acta 1997;1350:359—67. [36] Bohinski RJ, Di Lauro R, Whitsett JA. The lung-specific surfactant protein B gene promoter is a target for thyroid transcription factor 1 and hepatocyte nuclear factor 3, indicating common factors for organ-specific gene expression along the foregut axis. Mol Cell Biol 1994;14: 5671—81. [37] Strum N, Rossi G, Lantuejoul S, Papotti M, Frachon S, Claraz C, et al. Expression of thyroid transcription factor-1 in the spectrum of neuroendocrine cell lung proliferations with special interest in carcinoids. Hum Pathol 2002;33:175—82. [38] Rossi G, Cavazza A, Marchioni A, Migaldi M, Bavieri M, Facciolongo N, et al. Kit expression in small cell carcinomas of the lung: effects of chemotherapy. Mod Pathol 2003;16:1041—7.

233 [39] Araki K, Ishii G, Yokose T, Nagai K, Funai K, Kodama K, et al. Frequent overexpression of the c-kit protein in large cell neuroendocrine carcinoma of the lung. Lung Cancer 2003;40:173—80. [40] Micke P, Basrai M, Faldum A, Bittinger F, Ronnstrand L, Blaukat A, et al. Characterization of c-kit expression in small cell lung cancer: prognostic and therapeutic implications. Clin Cancer Res 2003;9:188—94. [41] Krystal GW, Honsawek S, Litz J, Buchdunger E. The selective tyrosine kinase inhibitor STI571 inhibits small cell lung cancer growth. Clin Cancer Res 2000;6:3319—26. [42] Zabarovsky ER, Lerman MI, Minna J. Tumor supressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers. Oncogene 2002;21:6915—35. [43] Wistuba II, Berry J, Behrens C, Maitra A, Shivapurkar N, Milchgrub S, et al. Molecular changes in the bronchial epithelium of patients with small cell lung cancer. Clin Cancer Res 2000;6:2604—10. [44] Braga E, Pugacheva E, Bazov I, Ermilova V, Kazubskaya T, Mazurenko N, et al. Comparative allelotyping of the short arm of human chromosome 3 in epithelial tumors of four different types. FEBS Lett 1999;454:215—9. [45] Ullmann R, Petzmann S, Sharma A, Cagle PT, Popper HH. Chromosomal aberrations in a series of large-cell neuroendocrine carcinomas: unexpected divergence from small-cell carcinoma of the lung. Hum Pathol 2001;32:1059—63. [46] Girard L, Zochbauer-Muller S, Virmani AK, Gazdar AF, Minna JD. Genome-wide allelotyping of lung cancer identifies new regions of allelic loss, differences between small cell lung cancer and non-small cell lung cancer, and loci clustering. Cancer Res 2000;60:4894—906. [47] Kok K, Naylor SL, Buys CH. Deletions of the short arm of chromosome 3 in solid tumors and the search for suppressor genes. Adv Cancer Res 1997;71:27—92. [48] Lindblad-Toh K, Tanenbaum DM, Daly MJ, Winchester E, Lui WO, Villapakkam A, et al. Loss-of-heterozygosity analysis of small-cell lung carcinomas using single-nucleotide polymorphism arrays. Nat Biotechnol 2000;18:1001—5. [49] Tomizawa Y, Nakajima T, Kohno T, Saito R, Yamaguchi N, Yokota J. Clinicopathological significance of Fhit protein expression in stage I non-small cell lung carcinoma. Cancer Res 1998;58:5478—83.