Phenotypic heterogeneity studied by immunohistochemistry and aneuploidy in non-small cell lung cancers

Non-small cell lung cancers (non-SCLC) differ from small cell lung cancers (SCLC) by many clinical features and prognosis. However, recent studies suggest that lung cancer heterogeneity frequently leads to the association of SCLC and non-SCLC in the same tumor. This phenotypic heterogeneity can be analyzed by immunohistochemistry using monoclonal antibodies (Mab) raised against differentiation related antigens. It may have clinical relevance inasmuch as the diversification of malignant cells is a well-known factor of tumor progression and may be due to chromosomal instability because inappropriate gene expression leads to the formation of antigens unrelated to cell lineage. Chromosomal instability in cancer leads to aneuploidy detectable by cell DNA content analysis. In a prospective study, we analyzed, in parallel, the expression of neuroendocrine related antigens by immunohistochemistry and the cell DNA content in frozen specimens from 40 patients who underwent complete surgical resection of primary non-SCLC in an attempt (a) to characterize the phenotypic heterogeneity and (b) to determine whether this heterogeneity is correlated with aneuploidy and clinical staging. Three Mabs were used in association as a marker of neuroendocrine antigen expression (S-L 11.14, MOC-1, and NE-25); reactivity of these Mabs in 9 SCLC and 3 lung carcinoid tissue sections was used as positive control. All SCLC and 2 of 3 lung carcinoids tested were homogeneously positive with Mabs S-L 11.14, MOC-1, and NE-25; 13 of 40 non-SCLC were homogeneously positive and 11 additional specimens focally positive with Mabs S-L 11.14, MOC-1, and NE-25. The frequency of this abnormal phenotype was significantly higher in poorly differentiated squamous cell carcinomas (chi 2 10.08; P less than 0.005), in clinical stage III non-SCLC (chi 2 5.93; P less than 0.02), and in tumors involving mediastinal lymph nodes (chi 2 5; P less than 0.03). The percentage of cells in the modal DNA of G0-G1 phase was significantly lower in non-SCLC homogeneously positive with Mabs S-L 11.14, MOC-1, and NE-25 [27.4 +/- 10.3% (SD)] in comparison with non-SCLC negative with these same Mabs [56.8 +/- 21.3%; P less than 0.01, Mann-Whitney U test]. We conclude that (a) mixed SCLC-non-SCLC differentiation is frequent and can be assessed by immunohistochemistry, (b) neuroendocrine differentiation in non-SCLC is mainly observed in poorly differentiated tumors and in advanced clinical stages, and that (c) this heterotopic phenotype is correlated with aneuploidy and has clinical implications.     

Expression of p64c-myc and neuroendocrine properties define three subclasses of small cell lung cancer

Twelve human small cell lung cancer (SCLC) cell lines and 6 non-SCLC cell lines were analysed with respect to expression of the c-myc, c-myb, and c-raf1 protooncogenes at the protein level. Analysis of p64c-myc protein expression in 12 SCLC cell lines resulted in the observation that it is present at high levels not only in cells with low, but also in those with moderate neuroendocrine differentiation. Neuroendocrine differentiation was based on parameters such as growth rate, colony formation, L-Dopa decarboxylase (DDC) activity, bombesin, and neurotensin described before. Surprisingly, in two cell lines with low neuroendocrine differentiation but without c-myc protein expression (SCLC-86M1 and NCI-H526) p75c-myb expression was observed which may therefore be able to substitute for the p64c-myc protein. Analysis of p74c-raf1 expression did not result in correlation with any growth or differentiation parameter since it was expressed at low levels in 11 out of 12 cases. We conclude that SCLC in vitro can be classified in three rather than two previously defined subclasses. In addition to the classic subclass with slow growth, high neuroendocrine differentiation, and absent or very low p64c-myc expression and the variant subclass with fast growth, absent to very low neuroendocrine differentiation, and high p64c-myc expression, we suggest a third subclass designated as transitional with moderate growth, moderate neuroendocrine differentiation, and high p64c-myc expression. Data on a small number of non-SCLC cell lines tested showed that high levels of p64c-myc correlate with high in vitro growth rates. This indicates that high p64c-myc levels may be associated with high proliferative activity, and lack of differentiation in lung cancer in general. The p74c-raf1 protein was found in all non-SCLC cell lines. Whether this classification of SCLC cell lines is applicable to SCLC in vivo remains to be determined.     

Coexpression of ganglioside antigen Fuc-GM1, neural-cell adhesion molecule, carcinoembryonic antigen, and carbohydrate tumor-associated antigen CA 50 in lung cancer.

With the aid of specific monoclonal antibodies, tumor tissues from 68 patients with lung cancer were examined for their expression of two small cell lung carcinoma (SCLC) antigens, Fuc-GM1 (fucosyl GM1; IV2FucII3NeuAc GgOse4) and neural-cell adhesion molecule (NCAM), and two broader tumor antigens, carcinoembryonic antigen (CEA) and carbohydrate cancer-associated antigen CA 50. Expression of Fuc-GM1 was seen in 75% and NCAM in 78% of the SCLC specimens, but also in 12 and 20% of non-SCLC. Either or both of these antigens were expressed in more than 90% of SCLC and in 25% of non-SCLC. CEA was found in more than 80% of SCLC and non-SCLC. Expression of CA 50 was seen in 65-68% of non-SCLC and SCLC, showing preference for SCLC and lung adenocarcinoma. In SCLC, cellular expression of Fuc-GM1 was generally seen together with NCAM and CA 50, but rarely with CEA. There was considerable inter- and intratumor heterogeneity in the expression of all four antigens. The results suggest that CEA is the antigen of choice for the detection of lung cancer regardless of histotype. In combined analysis of CEA, CA 50, Fuc-GM1 and NCAM, two patterns of antigen expression were recognized that appear to discriminate between SCLC and non-SCLC tumors, respectively. A considerable fraction of SCLC and non-SCLC tumors, however, exhibited similar patterns of antigen expression. The biological and clinical significance of these observations remains to be investigated.     

The Release of Chromogranin A and B Like Activity from Human Lung Cancer Cell Lines: A Potential Marker for a Subset of Small Cell Lung Cancer

Human small cell lung cancer (SCLC) is in vivo and in vitro characterized by a heterogeneous spectrum of neuroendocrine markers. the non-SCLC group is deprived of these markers, or expresses them in low quantities. in this paper we report on the release to the culture medium of neuroendocrine associated proteins, chromogranin A and B like activity (CABLA). the culture medium from three out of five SCLC cell lines and in onehive non-SCLC cell line contained significant levels of CABLA. Normal diploid foreskin fibroblasts and a histiocytic lymphoma cell line were deprived of CABLA production. the presence of CABLA in both SCLC and non-SCLC further stress their common histogenetic origin. the CABLA values were partly unrelated to other neuroendocrine markers. Determinations of CABLA could thus be a potential and valuable marker for a subset of SCLC.     

Demonstration of Chromogranin A in human neuroendocrine cell lines by immunohistology and immunoassay

We have used immunohistology and radioimmunoassay procedures to study Chromogranin A (CgA) in human neuroendocrine tumor cell lines, especially small cell lung cancers (SCLC). By immunohistology, CgA could be detected in 11 of 18 classical SCLC cell lines, in a medullary thyroid carcinoma (MTC) cell line, and in only one of 13 variant- or non-SCLC cell lines. By radioimmunoassay, CgA could be detected in the cells and culture media of all of the classical SCLC cell lines tested. Many of the classical SCLC cell lines also produced calcitonin (CT). These studies demonstrate that CgA production is a common feature of SCLC cell lines, especially those with neuroendocrine characteristics.     

Intermediate filament and cross-linked envelope expression in human lung tumor cell lines

Human lung tumor cell lines established from the major histological types of lung cancer were examined by immunofluorescent staining techniques for their patterns of intermediate filament (keratin, vimentin, and neurofilament triplet protein) expression. All cell lines examined, both small cell lung carcinoma (SCLC) and non-SCLC (squamous cell carcinoma, adenocarcinoma, large cell carcinoma, and mesothelioma) contained keratin, consistent with their epithelial derivation. These lung carcinoma cell lines also expressed vimentin, the characteristic intermediate filament of mesenchymal cells in vivo. In light of the proposed neuroectodermal origin of SCLC, cell lines were also studied for neurofilament expression. Two of four SCLC tumor cell lines, as well as non-SCLC cell lines, showed no reactivity with antibodies to neurofilament triplet protein. Two of the SCLC cell lines stained weakly with anti-neurofilament antibody. Examination of specific keratin patterns in human lung tumor cell lines by selective immunoprecipitation with keratin antiserum and sodium dodecyl sulfate-polyacrylamide gel electrophoresis indicated that small-sized keratin proteins (Mr 44,000 to 52,000) were present in cell lines derived from SCLC and non-SCLC types of lung cancer. Tumor cell lines exhibiting squamous differentiation by light microscopic criteria (i.e., intracellular keratin, intercellular bridging, "pearl" formation, and/or individual cell keratinization) also displayed a preponderance of intermediate-sized keratins (Mr 57,000 and 59,000) and exhibited another feature of terminal keratinocyte differentiation (cross-linked envelope formation). Mesothelioma cell lines had varying keratin profiles. The presence of keratin proteins in all SCLC cell lines examined argues against a neuroectodermal origin for these tumors and is consistent with the notion that these tumors arise from a common bronchial "stem cell," similar to that from which other types of bronchogenic carcinomas arise.     

Immunophenotype of small cell lung carcinoma. Expression of NKH-1 and transferrin receptor and absence of most myeloid antigens

NKH-1 is a monoclonal antibody that reacts with human natural killer (NK) cells and neural tissue. Because other monoclonal antibodies reacting with NK cells have been found on small cell lung carcinoma (SCLC), frozen tissue sections of 22 lung tumors including nine SCLC, two bronchial carcinoids, and 11 non-SCLC were tested for the presence of NKH-1 antigen by a sensitive alkaline phosphatase/anti-alkaline phosphatase technique. The labeling reactions of NKH-1 in frozen tissue sections were compared with reactions of a panel of 21 other monoclonal antibodies against NK cells, leukocyte antigens, cytokeratins, or nonlineage specific antigens. The antibody NKH-1 reacted strongly and diffusely with all of the SCLC and bronchial carcinoids but with none of the non-SCLC. NKH-1 also strongly labeled peripheral nerves in tissues adjacent to tumor. Two antibodies to cytokeratins reacted with all of the tumors and outlined tumor cells well, distinguishing them from surrounding stromal cells and leukocytes. OKT9, an antibody against transferrin receptor labeled all SCLC and eight of 11 non-SCLC but did not react with bronchial carcinoid. The antibodies Leu-M1, OKT10, Leu-7, and My4 reacted with 67%, 33%, 22%, and 11%, respectively, of the SCLC tested. The remaining 14 antibodies, including several with leukocyte specificity, labeled neither SCLC nor bronchial carcinoid. Thus, SCLC has a distinct immunophenotype (NKH-1 positive, keratin positive, and transferrin receptor positive), which may be helpful distinguishing this tumor from other tumors of lung including non-SCLC. SCLC infrequently expresses other leukocyte-associated antigens.     

Determination of various tumor markers, with special reference to neuron-specific enolase in lung cancer

Serum neuron-specific enolase (NSE) was measured in 23 patients with small cell lung cancer (SCLC) and 184 patients with non-small cell lung cancer (non-SCLC), both of which were untreated. Increased levels of serum NSE were observed in 82.6% (19/23) of SCLC, whereas 9.8% (18/184) of non-SCLC had positive results showing an overall positive rate of 17.9% (37/207) in lung cancer cases. In addition, the elevation of serum NSE levels in non-SCLC patients seemed to suggest poor prognosis. Elevated serum NSE levels returned to normal with either surgical resection of the tumor or response to chemotherapy, after which serum NSE levels were again raised to levels higher than the previous ones in cases of relapse or progression. The evaluation of serum NSE may be a useful marker for both diagnosis and monitoring of responsiveness to therapy as well as for recognition of relapse and progression in SCLC. Identification of NSE as assessed by immunohistochemical procedure employing the ABC method on formalin-fixed paraffin-embedded tissue sections in lung cancer cases of each histological type, showed that some materials from non-SCLC cases were positively stained despite the presence of normal serum NSE levels, and did not always parallel the serum levels. Among other various tumor markers determined, serum CA 19-9 had a relatively high positive rate of 38.2% (42/110) in adenocarcinoma of the lung.     

Biological and Clinical Implication of Neuron-Specific Enolase and Creatine Kinase BB in Small Cell Lung Cancer

The specificity of neuron-specific enolase (NSE) and creatinekinase BB (CK-BB) for small cell lung cancer (SCLC) was determinedby biological and immunohistochemical procedures in lung cancertissues and cultured cell lines. Average values of extractableNSE and CK-BB of SCLC tissues were significantly higher thanthose of non-SCLC and normal lung tissues. A large amount ofNSE and CK-BB was demonstrated in SCLC cell lines. Immunohistochemical examination showed positive staining forNSE and CK-BB in most cases of SCLC and in a few cases of non-SCLC.From these data NSE and CK-BB should be considered to be highlyspecific for SCLC. In a clinical study serum values exceeding 10 ng/ml for NSEand 1.5 ng/ml for CK-BB were set as positive for the enzymes.Positive rates in SCLC were 71.4% for NSE and 65.3% for CK-BB,which were significantly higher than those in non-SCLC. Allpositive cases were in an advanced stage. Consecutive dailyNSE determinations during induction chemotherapy showed transientelevation immediately after the initiation of drug administration(tumor lysis syndrome), followed by a decline to the normalrange in responders. This phe nomenon seems to indicate tumorsensitivity to cytotoxic drugs. NSE positive non-SCLC was assensitive to cytotoxic drugs as SCLC. These findings indicatethat lung cancer with elevated serum NSE and CK-BB levels atdiagnosis should be strongly suspected of being SCLC in theadvanced stage.     

Electron microscopy in small cell lung carcinomas: clinical correlation

In order to investigate the relationship of subcellular differentiation of small cell lung carcinomas (SCLC) and clinical response, we reviewed the electron microscopic (EM) features of tumor biopsy specimens from 33 patients with SCLC diagnosed by light microscopy (LM). These tumors were divided by EM into four groups according to the ultrastructural features. Group I (13 patients) had tumors with only neurosecretory granules on EM. Group II (11 patients) had tumors with neurosecretory granules and other subcellular features of non-SCLC. Group III (five patients) had tumors that lacked neurosecretory granules but contained subcellular features of non-SCLC. Group IV (four patients) had tumors that lacked all of these features. The complete and partial response rate to systemic chemotherapy with or without radiation therapy was 88% in the total population studied. The response rates were not statistically different in any of the four groups based on EM findings. The results of this study suggest that the LM diagnosis of SCLC alone adequately identifies lung cancer patients with a high response rate to systemic therapy.     

Oxytocin receptor pattern of expression in primary lung cancer and in normal human lung

In order to assess if oxytocin- and vasopressin-induced mitogenic effects detected on small-cell lung carcinoma (SCLC) cell lines could be transposed on primary SCLC, the aim of the present work was to identify mediators of these mitogenic actions on primary tumours samples. This was addressed on normal human lung tissue, on SCLC and on non-SCLC (NSCLC). Herein, we observe, in normal human lung, that OTR is colocalized with vascular endothelial cells of the lung and is not expressed by lung cells of epithelial nature. We detected mRNA amplification of V1aR, V2R and of a V2R variant. We observed that 86% of SCLC biopsies analyzed expressed at least the OTR and that 71% expressed the OTR, the V1aR and the V2R altogether. Comparatively, 50% of NSCLC biopsies tested expressed at least the OTR and 32% expressed the OTR, the V1aR and the V2R altogether. The occurrence of the V1bR/V3R is of 28 and 18% for SCLC and NSCLC, respectively. Nevertheless, for the SCLC biopsies analyzed in this study, V1bR/V3R expression correlates, in all cases, with the expression of all the other neurohypophysial peptide receptors. Our results suggest that neurohypophysial peptide antagonists may offer promise as a potential new therapeutic modality for the treatment of lung cancer expressing at least one of the neurhypophysial peptide receptor subtypes.     

Association between histological type and neuroendocrine differentiation on drug sensitivity of lung cancer cell lines.

Because most small-cell lung cancers (SCLC) are initially chemosensitive and express neuroendocrine (NE) cell markers and most non-SCLC tumors (NSCLC) are chemoresistant and do not express NE cell markers, we investigated the association between morphological type and NE cell differentiation with in vitro chemosensitivity. We tested a panel of 55 lung cancer cell lines established from previously untreated patients. These were tested against five cytotoxic drugs commonly used in the therapy of lung cancer, using the MTT assay. For comparative purposes, we also tested cell lines established from previously treated patients with SCLC and from colorectal tumors. The logarithms of the IC50 values of all of the cell lines were normally distributed, permitting the use of Student's t-test for assessment of differences. In general, the in vitro sensitivities of SCLC, NSCLC, and colorectal cell lines mirrored the clinical experience with these tumor types. Cell lines started from previously treated patients with SCLC were more resistant than those from previously untreated patients who responded to initial therapy. For all of the cell lines, the sensitivities to the five drugs tested were highly significantly correlated with each other. Thus, for comparative purposes, each group could be assigned an average standardized mean rank. About 15% of NSCLC tumors express multiple neuroendocrine (NE) cell markers and 4 of 5 lines from these NSCLC-NE tumors were relatively chemosensitive, similar to SCLC lines and significantly different from other NSCLC lines. Other NE cell lines tested included bronchial carcinoids and cell lines from small-cell carcinomas arising in extra-pulmonary locations (ExPuSC).(ABSTRACT TRUNCATED AT 250 WORDS)     

Analysis on the Characteristics and Prognosis of Pulmonary Neuroendocrine Tumors

Objective: To retrospectively review the clinical characteristics and analyze the prognostic factors of Chinesepatients with pulmonary neuroendocrine tumors. Materials and Methods: The clinical data of 176 patientswith pulmonary neuroendocrine tumors in Chinese PLA General Hospital from Mar., 2000 to Oct., 2012 wereretrospectively analyzed. The parameters were evaluated by univariate and multivariate analysis, including thegender, age, smoking history, family history, TNM staging, localization (central or peripheral), tumor size, nodalstatus, histological subtype and treatment (operation or non-operation). Results: There were 23 patients withtypical carcinoids (TC) (13.1%), 41 with atypical carcinoids (AC) (23.3%), 10 with large cell neuroendocrinecarcinoma (LCNEC) (5.7%) and 102 with small cell lung cancer (SCLC) (57.9%). The median follow-up time was64.5 months for AC, 38 months for LCNEC and 27 months for SCLC. The typical carcinoid censored data was 18(more than 50% of the patients), so the median follow-up time was not obtained, and actuarial 5-year survivalsfor TC, AC, LCNEC and SCLC were 75.1%, 51.7%, 26.7% and 38.8%, respectively. COX univariate analysisrevealed that the age (P=0.001), histological subtype (P=0.005), nodal status (P=0.000), treatment (P=0.000) andTNM staging (P=0.000) were the prognostic factors of the patients with pulmonary neuroendocrine tumors,whereas its multivariate analysis showed that only the age(P=0.001), TNM staging (P=0.002) and treatment(P=0.000) were independent prognostic factors. Conclusions: Radical surgery remains the treatment of choice,and is the only curative option. The age, TNM staging and treatment are confirmed to be the independentprognostic factors in multivariable models for pulmonary neuroendocrine tumors.     

Differential inactivation of caspase-8 in lung cancers

Caspase-8 (CASP8) is an apoptosis inducing cysteine protease which is activated through the formation of a death-inducing signaling complex when death receptors are complexed to their specific ligands. Recent reports indicate that CASP8 expression is lost via a combination of promoter methylation and allelic loss in subset of neuroblastomas. We investigated the state of the gene in lung tumors and cell lines. RT-PCR studies indicated that gene expression was lost in most (27 of 34, 79%) of small cell lung carcinoma (SCLC) cell lines, but expression was retained in all 22 non-SCLC (NSCLC) lines tested. Loss of gene expression at the RNA level was associated with absent protein expression by Western blotting and lack of CASP8 enzymatic activity. Methylation of the promoter region of the CASP8 gene was present in 16 of 27 (59%) of the SCLC lines lacking gene expression. All methylated cell lines lacked the presence of an unmethylated allele indicating biallelic methylation or loss of non-methylated allele. Promoter methylation was absent in all SCLC and NSCLC cell lines retaining gene expression, and all of these lines had the unmethylated form of the gene. One non-expressing SCLC cell line, NCI-H82, had a homozygous deletion at 2q33 encompassing the chromosomal location of the CASP8 gene. The mechanism of gene inactivation in the remaining 10 of 27 (37%) non-expressing SCLC cell lines is unknown. Using five polymorphic markers for 2q33 a high frequency of allelic loss was present in SCLC lines. Analyses of fresh tumors showed that 15 of 43 (35%) of the SCLC, seven of 40 (18%) of bronchial carcinoids and none of 44 NSCLC tumors had CASP8 promoter methylation. Because only approximately 60% of SCLC cell lines lacking CASP8 expression were methylated, extrapolating from the cell line data, we estimate that approximately 58% of SCLC and 30% of bronchial carcinoids lack CASP8 expression. Thus, CASP8 expression is absent in a subset of both high grade (SCLC) and low grade (carcinoid) neuroendocrine lung tumors but not in NSCLC, which usually lack neuroendocrine features. CASP8 may function as a tumor suppressor gene in neuroendocrine lung tumors.     

Immunocytochemical detection of human lung cancer heterogeneity using antibodies to epithelial, neuronal, and neuroendocrine antigens

Lung cancers were investigated for their heterogeneity as expressed by their immunoreactivity for cytokeratins and neurofilament proteins, as well as for the neuroendocrine differentiation antigen MOC-1. Using broadly cross-reacting antibodies, cytokeratins were detected in nearly all cases of lung carcinomas. Keratinization could be detected only in cases of moderately to well-differentiated squamous cell carcinoma (SQC) using a monoclonal antibody to cytokeratin 10, while a monoclonal antibody reactive with cytokeratin 18, and specific for glandular epithelia, reacted with adenocarcinomas, small cell lung carcinomas (SCLC), and lung carcinoids. In SQC this antibody could detect non-squamous cell differentiation, showing increasing numbers of positive cells with decrease of histologically detectable SQC differentiation. Cells positive for neurofilaments were demonstrated in some of the poorly differentiated SQCs and in some of the cases of SCLC, possibly representing the variant type of SCLC. Also in some of the lung carcinoids neurofilament proteins were present, colocalizing with cytokeratins. MOC-1 was present in all SCLC and lung carcinoids. This antibody could also detect neuroendocrine differentiation in all combined small cell carcinomas, in one poorly differentiated adenocarcinoma, and in about 30% of the poorly differentiated SQCs. Therefore, lung cancer heterogeneity can be detected using a panel of well-defined antibodies to intermediate filaments in combination with the MOC-1 antibody. The use of these antibodies in diagnosis can have prognostic significance and can lead to a more selective therapeutic approach.     

Quantitative distribution of cluster 1 small cell lung cancer antigen in cancerous and non-cancerous tissues, cultured cells and sera

Three monoclonal antibodies (mAbs), NCC-LU-243, -244 and -246, detected three different epitopes on a 145-kDa cell membrane antigen, which had been designated as the cluster 1 antigen at the First International Workshop on Small Cell Lung Cancer (SCLC) Antigens. The distribution of the antigen in various tissues, cultured cells and sera was examined by immunohistochemistry and sandwich radioimmunoassay using these mAbs. The antigen is a normal differentiation antigen and is present in neuronal, neuroendocrine and cardiac muscle cells. The level of the antigen was highest in central nervous tissues, while it was undetectable in the liver, kidney and peripheral lung. Among tumor tissues, the antigen was detected only in SCLC, carcinoid tumor and neuroblastoma, indicating its usefulness as a marker for discriminating SCLC from non-SCLC. The level of the antigen varied among SCLC tissues and tended to be lower in variant-type cultured SCLC cells. Although an increase in the antigen level was observed in sera of some patients with advanced SCLC, the antigen did not possess any additional value over neuron-specific enolase as a serum tumor marker for monitoring SCLC patients.