Lung Cancer Biopsy Specimens Research Articles

Expert consensus on the pathological diagnosis of small biopsy specimens of lung cancer

肺癌患者小活检标本病理诊断至关重要，尤其是对于晚期肺癌患者，其精准治疗取决于准确的肺癌病理类型及相应的分子病理检测。本共识结合第5版WHO肺癌组织学分类、相关临床肺癌诊治指南或共识要求及有关临床病理研究进展，对肺癌小活检标本诊断术语、病理亚型诊断注意事项、标志物使用及其特点、诊断局限性等方面进行说明或推荐，旨在加强肺癌小活检标本组织病理诊断的同质性及规范性，进一步提高肺癌亚型诊断准确性。.

Comparative study of deep learning models on the images of biopsy specimens for diagnosis of lung cancer treatment

ObjectiveTo compare the application value of different traditional deep learning models in diagnosing and classifying lung cancer. MethodsAccording to the biopsy samples of our hospital from January 2018 to November 2022, 37 patients treated in this department were selected as the study subjects. Nonsmall and small cell lung cancer biopsy specimens were obtained and stained. Two experienced pathologists diagnosed the biopsy specimens. Multiple in-depth learning methods were used to distinguish between cancer and noncancer biopsies. In this study, we compared the application value of traditional deep learning models in lung cancer diagnosis and Classification. ResultsThe study tested several popular CNN architectures based on image block classification: AlexNet, VGG, ResNet, and SqueezeNet, comparing two types of training schemes: training from scratch and fine-tuning the entire pretrained network. The AUC of the deep learning model is more reasonable (0. 8808–0. 9121). Except for Resnet-50, the AUC of the training is higher than that of the fine-tuning of the whole network. ConclusionDeep learning analysis can accelerate the detection speed of whole section images (WSI) and maintain a similar detection rate with pathologists.

Proportion of biopsy specimens containing a tumor when compared to all biopsy specimens by transbronchial biopsy.

The lung cancer biopsy specimens obtained by endobronchial ultrasound-guide sheath (EBUS-GS) trans lung biopsy occasionally do not contain cancer cells. It is a problem that there is a possibility that they may not contain cancer cells. To investigate the proportion of biopsy specimens containing cancer cells in total biopsy specimens. Patients with lung cancer diagnosed by EBUS-GS were selected. The primary end point was the proportion of specimens containing tumors in the total specimens obtained by EBUS-GS. Twenty-six patients were investigated. The percentage of specimens containing cancer cells in the total specimens was 79.0%. The proportion of biopsy specimens containing cancer cell to all biopsy specimens by EBUS-GS was high, but not 100%.

Novel Robotic-Assisted Cryobiopsy for Peripheral Pulmonary Lesions

PurposeTissue acquisition in lung cancer is vital for multiple reasons. Primary reasons reported for molecular testing failure in lung cancer biopsy specimens include insufficient amount of tumor cells provided and inadequate tissue quality. Robotic bronchoscopy is a new tool enabling peripheral pulmonary lesion sampling; however, diagnostic yield remains imperfect possibly due to the location of nodules adjacent to or outside of the airway. The 1.1-mm cryoprobe is a novel diagnostic tool and accesses tissue in a 360-degree manner, thus potentially sampling eccentric/adjacent lesions. This study examines the diagnostic yield of the cryoprobe compared to standard needle aspiration and forceps biopsy. It additionally evaluates yield for molecular markers in cases of lung cancer.MethodsThis is a retrospective analysis of 112 patients with 120 peripheral pulmonary lesions biopsied via robotic bronchoscopy using needle aspirate, forceps, and cryobiopsy.ResultsThe overall diagnostic yield was 90%. Nearly 18% of diagnoses were made exclusively from the cryobiopsy sample. Molecular analysis was adequate on all cryobiopsy samples sent. Digital imaging software confirmed an increase in quantity and quality of samples taken via cryobiopsy compared to needle aspirate and traditional forceps biopsy.ConclusionUsing the 1.1-mm cryoprobe to biopsy PPN combined with the Ion robotic bronchoscopy system is safe, feasible, and provides more diagnostic tissue than needle aspirates or traditional forceps biopsies. The combination of cryobiopsy with robotic-assisted bronchoscopy increased diagnostic yield, likely due to its 360-degree tissue acquisition which is beneficial when targeting extraluminal lesions adjacent to the airway.

Epigenetic regulation of lung cancer cell proliferation and migration by the chromatin remodeling protein BRG1

Malignant lung cancer cells are characterized by uncontrolled proliferation and migration. Aberrant lung cancer cell proliferation and migration are programmed by altered cancer transcriptome. The underlying epigenetic mechanism is unclear. Here we report that expression levels of BRG1, a chromatin remodeling protein, were significantly up-regulated in human lung cancer biopsy specimens of higher malignancy grades compared to those of lower grades. Small interfering RNA mediated depletion or pharmaceutical inhibition of BRG1 suppressed proliferation and migration of lung cancer cells. BRG1 depletion or inhibition was paralleled by down-regulation of cyclin B1 (CCNB1) and latent TGF-β binding protein 2 (LTBP2) in lung cancer cells. Further analysis revealed that BRG1 directly bound to the CCNB1 promoter to activate transcription in response to hypoxia stimulation by interacting with E2F1. On the other hand, BRG1 interacted with Sp1 to activate LTBP2 transcription. Mechanistically, BRG1 regulated CCNB1 and LTBP2 transcription by altering histone modifications on target promoters. Specifically, BRG1 recruited KDM3A, a histone H3K9 demethylase, to remove dimethyl H3K9 from target gene promoters thereby activating transcription. KDM3A knockdown achieved equivalent effects as BRG1 silencing by diminishing lung cancer proliferation and migration. Of interest, BRG1 directly activated KDM3A transcription by forming a complex with HIF-1α. In conclusion, our data unveil a novel epigenetic mechanism whereby malignant lung cancer cells acquired heightened ability to proliferate and migrate. Targeting BRG1 may yield effective interventional strategies against malignant lung cancers.

Novel multiplex droplet digital PCR assay for scoring PD-L1 in non-small cell lung cancer biopsy specimens

ObjectivesImmune checkpoint inhibitors have become integrated into the clinical management of non-small cell lung cancer (NSCLC). Using RTqPCR, we have previously identified a gene expression panel that detected presence of malignant cells (MMP9:TIMP3 ratio) and quantified PD-L1 transcript levels in small biopsy specimens. However, RTqPCR has diagnostic limitations as it does not generate absolute copy number and is not readily multiplexed. To address this, we have developed a multiplex droplet digital PCR (ddPCR) assay. Materials and methodsBiopsies obtained from NSCLC patients (n = 48 adenocarcinoma and n = 40 squamous cell carcinoma) and control lung biopsy specimens (n = 20) were analysed. Absolute MMP9, TIMP3 and PD-L1 transcript copy numbers were determined within a single assay by multiplex ddPCR using Taqman primers and the QX200 Droplet Digital PCR System. Results and conclusionsUsing our optimised triplex ddPCR assay, the MMP9:TIMP3 ratio was significantly elevated in NSCLC biopsies and using a cut-off of >0.028, was 99% (95% CI; 80.5-94.5) sensitive and 80% specific for identifying malignant biopsies. The PD-L1:TIMP3 ratio significantly associated with PD-L1 tumour cell immunohistochemistry staining (r = 0.539, p < 0.0001) and was significantly higher in biopsies with >50% PD-L1 tumour cell staining (p < 0.0001). In summary, a major advantage of our workflow is that it can accurately quantify PD-L1 tumour levels and provide sufficient nucleic acid for screening additional targetable mutations such as EGFR, ALK and ROS1 from a single small biopsy, thereby potentially avoiding the need for re-biopsy. Future studies will need to determine diagnostic ddPCR values that are predictive of clinical response to PD-1/PD-L1 immunotherapy.

The feasibility of using small biopsy samples from lung cancer for targeted next-generation sequencing.

e20584 Background: As the genomic based personalized medicine has been advanced, biomarker analysis using targeted next-generation sequencing (NGS) becomes crucial diagnostic approach for treatment of lung cancer. Most patients of lung cancer are diagnosed on small biopsy specimen, but the feasibility of the small tissue samples for targeted NGS has not been evaluated. We aimed to investigate the feasibility of using small tissue samples of lung cancer for targeted NGS. Methods: The medical records of the patients who underwent biopsy and targeted NGS were reviewed. Two types of panel for targeted NGS were used which were AmpliSeq Cancer Hotspot panel v2 and Oncomine Focus Assay. Results: A total of 603 biopsy specimens of lung cancer were analyzed from April 2014 to October 2016. The sites of biopsy were lung or bronchus (51.2%), lymph node (26.2%), Pleura (2.3%), liver (2.5%), Brain (1.7%) and the others (adrenal gland, bone, muscle, spleen, stomach, thymus). Percutaneous transthoracic core needle biopsy (PTCNB) was most frequently used method for acquisition of tissue (24.9%). The other methods of biopsy were bronchoscopy (14.4%), endobronchial ultrasound (EBUS, 23.8%), Video-assisted transthoracic surgery (VATS, 9.9%) and so on. The types of tissue samples were formalin-fixed, paraffin-embedded (FFPE, 80.6%) and fresh tissue (19.4%). All of these samples, 398 cases (66.0%) were successfully analyzed using targeted, NGS. In 34% of analysis failure cases, 23.5% were not adequate for DNA or RNA extraction and 10.5% had amount not to be enough. On multivariate analysis, PTCNB was more likely to fail than other methods of biopsy such as bronchoscopy, EBUS (HR 2.31, 95% CI 1.4-3.8, p = 0.001). Conclusions: The biopsy of lung cancer for targeted NGS was taken from a variety of sites via different methods, and the type of tissue samples was mostly FFPE. Considering the yield of our study, small biopsy specimens could be used to investigate genomic characteristics of lung cancer via targeted NGS.

P1.01-057 Metastatic Lung Cancer at a Tertiary Cancer Centre in South India

Nonsmall cell lung cancer (NSCLC) has varying epidemiological patterns in different countries and also in different regions of each country. In a country with a high prevalence of lung cancer such as India, regional variations in demography exist. We did a retrospective analysis of histologically confirmed metastatic NSCLC patients who presented to our Department of Medical Oncology between August 2012 and July 2014. The patients were interviewed regarding their history of smoking (never smokers, light smokers, and heavy smokers). Never smokers were defined as those who have smoked <100 bidi/cigarettes in their life until disease onset. Light smokers were defined as those who smoke <10–100 bidi/cigarette pack years. Heavy smokers were defined as those who smoke more than 100 bidi/ cigarette pack years until disease. All lung cancer biopsy specimens were histologically characterized by morphology and immunohistochemistry (IHC). A diagnosis of AC was made if IHC staining was positive for CK7, TTF1, and napsin A; similarly, SCC was diagnosed if CK5/6 and p63 staining was positive. After confirmation of the histology subtype, patients were confirmed to have metastatic disease after assessment with contrast-enhanced computerized tomography thorax, abdomen scan, and radiolabeled bone scan. Our study did not include the small cell lung cancer patients. The data were analyzed on SPSS version 22 (IBM) software. A total of 304 patients were analyzed. About 55.6% of the patients were in the age group of 41–60 years. About 79.6% of the patients were symptomatic for <6 months before presentation. About 63.5% of the patients were smokers presenting with a median age of 59 years whereas nonsmokers formed 36.51% of the patients presenting with a median age of 47 (P < 0.001). About 82.6% of the male patients and 4.1% of female patients were smokers. Equal number of all patients had adenocarcinoma (AC) and squamous cell carcinoma (SCC) histology. AC histology was more common in the nonsmoking group (62% of patients). SCC histology was seen in 54.3% of smokers. Metastasis to the contralateral lung and pleura was seen in 58.2% of patients. NSCLC presents at a young age. Smoking is a significant risk factor and it is common in the urban populations as in the rural areas. Both AC and SCC histologies presented in equal proportions.

QUANTITATIVE ASSESSMENT OF BETA-III TUBULIN EXPRESSION IN LUNG ADENOCARCINOMA AND SQUAMOUS CELL CARCINOMA

Background. Beta-III tubulin (TUBB3) is a tumor-specific isoform of the microtubule protein beta-tubulin. TUBB3 is considered to be a marker of adverse prognosis and tumor resistance to therapy with taxanes and Vinka alkaloids. Association between TUBB3 expression and histological type of the tumor has not been studied properly yet. Objective. The expression level of TUBB3 in non-small cell lung cancer biopsy specimens has been measured on 2 groups of patients with adenocarcinoma and squamous cell carcinoma. Materials and methods. The samples with adenocarcinoma (n = 43) and squamous cell carcinoma (n = 39) were converted to suspension, filtered, fixed with 4 % formaldehyde, stained with monoclonal antibodies for TUBB3 and DyLight 650-conjugated secondary antibodies to mouse IgG and analyzed by flow-cytometry. The method was developed in N.N. Blokhin Russian Cancer Research Center. Results. The average level of TUBB3 expression in the adenocarcinoma group is higher than in the squamous cell carcinoma group (33.1 ± 12.4 % of cells expressing the marker in adenocarcinoma vs. 26.0 ± 13.6 % in squamous cell carcinoma; differences were statistically significant). The average level of TUBB3 expression in adenocarcinoma is 29.7 ± 8.1 % in female and is 34.9 ± 13.9 % in male (differences statistically insignificant). Since the group of patients with adenocarcinoma was presented by both men and women, while the group of patients with squamous cell carcinoma was only men, from the analysis was excluded all female patients. Differences between the groups remained statistically significant. Conclusion. TUBB3 expression in tumor tissue does not depend on the gender, at least among patients with adenocarcinoma of the lung, and at the same time, the level of TUBB3 in adenocarcinoma tissue is higher in comparison with squamous cell carcinoma tissue.

Significance and evaluation of anaplastic lymphoma kinase by immunohistochemistry in non-small cell lung cancer.

We used immunohistochemistry and reverse transcription polymerase chain reaction (RT-PCR) to evaluate anaplastic lymphoma kinase (ALK) protein expression and gene rearrangements, respectively, in 283 cases of wild-type epidermal growth factor receptor (EGFR) non-small cell lung cancer biopsy specimens. Immunohistochemistry was positive for ALK in 52 cases (18.4%), and there was no significant difference in staining between various monoclonal antibodies (Roche ALK test kit, D5F3, p-ALK, and EML4-ALK). On RT-PCR, 36 cases (12.7%) were positive for ALK. Immunohistochemistry and RT-PCR were both positive in 35 cases and both negative in 230 cases, and both have a high consistency (265/283, 93.6%). Including 17 cases, immunohistochemistry was positive but RT-PCR was negative, and in one case, immunohistochemistry was negative but RT-PCR was positive. On fluorescence in situ hybridization (FISH) testing of these 18 cases, only three cases were positive (one RT-PCR was positive; two immunohistochemistry were positive). There is a high prevalence of ALK positivity in wild-type EGFR non-small cell lung cancer. Immunohistochemistry for the detection of ALK gene rearrangements was highly consistent with RT-PCR, and thus, it is a good screening tool but produces false positive results that necessitate further screening by RT-PCR or FISH.

Abstract 4539: Nitroxide-aspirin conjugates: A new class of NSAIDs

Abstract Novel nitroxide-aspirin conjugates were synthesized and tested using human lung cancer cell lines. Previously, cyclooxygenase (COX2) was found to be elevated in lung cancer biopsy specimens relative to normal tissue (He et al., 2012; PlosOne 7:e31087). Lung cancer cellular proliferation is inhibited by nonsteroidal anti-inflamatory drugs (NSAIDs) such as indomethacin and aspirin (Hida et al., 1998; Anticancer Res. 18: 775). Dithiolethione-valproate conjugates inhibit COX-2 activity, increase E-cadherin expression and decrease the proliferation of non-small cell lung cancer (NSCLC) cells (Moody et al., 2010, Lung Cancer 68: 154). Here the effects of nitroxide-aspirin conjugates were investigated on NSCLC cells. 1KT121A, a 4-carboxy Tempo-aspirin conjugate, inhibited the proliferation of NCI-H1299 or A549 cells with IC50 values of 125 uM using the MTT assay whereas aspirin and tempol had IC50 values of 750 and &amp;gt;1000 uM, respectively. 1KT121A, which has an ester link, was approximately 5-fold more potent at inhibiting NSCLC cellular proliferation than aspirin plus tempol or 2KT106A, which is a Tempo-aspirin conjugate with an amide link. 1KT121A was more potent than 2KT106A or tempol plus aspirin at reducing prostaglandin E2 levels in A549 cells. Similar results were obtained if the nitroxide moiety was changed from 4-carboxy Tempo to 3-carboxy proxyl or 5-carboxy-1,1,3,3-tetramethylisoindolin-yloxyl (CTMIO). 1KT121A potentiated the ability of gefitinib, an EGF receptor tyrosine kinase inhibitor, to kill NSCLC cells. Because 1KT121A increased E-cadherin and reduced vimentin, it may impair epithelial-mesenchymal transitions of NSCLC cells. All of the nitoxide analogues were antioxidants based on the ability to inhibit reactive oxygen species in NSCLC cells treated with hydrogen peroxide. Nitroxide-aspirin conjugates represent a novel class of NSAIDs which are antioxidants that inhibit NSCLC proliferation. Citation Format: Terry W. Moody, Robert T. Jensen, Komba Thomas, Kathryn E. Fairfull-Smith, Steven E. Bottle, Lisa Ridnour, David A. Wink. Nitroxide-aspirin conjugates: A new class of NSAIDs. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4539. doi:10.1158/1538-7445.AM2015-4539

CXCL1 Regulation in Human Pulmonary Epithelial Cells by Tumor Necrosis Factor

Background/Aims: The chemokine CXCL1 has been reported to be expressed in lung airway epithelium and non-small cell lung cancer biopsy specimens. In this study, we investigated the effects of TNF-α, an abundant cytokine detected in inflammation and various cancers, on CXCL1 release by human A549 lung carcinoma epithelial cells. Methods: CXCL1 expression was determined by ELISA and RT-PCR. TNF-α signaling was examined by western blotting. Monocyte migration was assayed by a Transwell migration system. Results: TNF-α stimulated CXCL1 release and mRNA expression, and this release was inhibited by inhibitors of JNK, p38 MAPK, PI-3K/Akt and AP-1 transcription factor. TNF-α treatment was followed by JNK, p38 MAPK and PI3K/Akt activation. However, only the JNK inhibitor could reduce the CXCL1 mRNA level, suggesting that JNK is required mainly for CXCL1 mRNA synthesis, whereas p38 MAPK and PI-3K/Akt might be responsible for CXCL1 secretion. Dexamethasone (dex) and TGF-β reduced CXCL1 secretion, with dex upregulating the expression of MAP kinase phosphatase-1 and TGF-β causing smad2/3 activation and nuclear translocation. A functional analysis showed that the released CXCL1 enhanced monocyte migration and could be abolished by a CXCL1 neutralizing antibody and CXCR antagonist. Conclusion: We demonstrate that TNF-α induces CXCL1 expression through the JNK, p38 MAPK and PI-3K/Akt signaling pathways in human pulmonary epithelial cells.

Abstract 4696: Expression of the DNA repair protein UNG predicts lung cancer sensitivity to pemetrexed

Abstract Uracil misincorporation is a direct consequence of TS inhibition by the multi-target antifolate, pemetrexed. The DNA base excision repair (BER) enzyme Uracil DNA glycosylase (UNG) is the major glycosylase responsible for the removal of misincorporated genomic uracil. We have previously illustrated that human colon cancer cells (DLD1) with engineered stable somatic knockout of UNG expression are hypersensitive to pemetrexed owing to the cytotoxicity of genomic uracil accumulation. Here, we examined the relationship between the expression of UNG and the pemetrexed sensitivity in a small panel of lung cancer cell lines (n∼10). The levels of UNG were measured by western blot (for protein) and RT-PCR (for mRNA). Results revealed that UNG mRNA expression is well correlated with protein expression in all cell lines. Moreover, there was a wide range of distribution of UNG levels in the lung cancer cell lines evaluated. UNG cutting assay and AP site measurements confirm that these differences in UNG expression result in varied potential for BER activation and subsequent AP site formation. To determine the role of this variation in UNG expression in pemetrexed cytotoxicity, colony survival studies were performed. In general, cell lines with higher UNG expression were more resistant to pemetrexed. H460 cells having 7-fold lower expression of UNG mRNA than H1975 cells are &amp;gt;5 fold more sensitive to pemetrexed than H1975 cells (H460 IC50 = 200nM; H1975 IC50 = &amp;gt;1000nM). Analysis of tissue microarray data in Oncomine confirms variations in UNG expression in primary human lung cancer samples. Moreover, small cell lung cancer and squamous cell lung cancer, both notoriously resistant to pemetrexed, have significantly increased median UNG expression compared to all other lung cancer subtypes (1.5-fold greater UNG expression p&amp;lt;0.005). These data motivate future evaluation of UNG as a clinical predictor of pemetrexed response using lung cancer biopsy specimens. Together, our results clearly indicate that the BER protein, UNG is differentially expressed in lung cancer cells. The inherent variation in UNG expression may have particular impact on our ability to predict the therapeutic efficacy of pemetrexed in lung cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4696. doi:1538-7445.AM2012-4696

Tissue Banking of Diagnostic Lung Cancer Biopsies for Extraction of High Quality RNA

There is a clear need to develop a practical approach to obtain high quality RNA for gene expression analysis from lung cancer patients. Current approaches are restricted to using material from surgical resection specimens. We systematically investigated whether high quality RNA could be obtained from routine lung cancer diagnostic biopsies to determine the optimum method. Extra biopsies were taken at diagnosis from patients later confirmed to have lung cancer. Comparisons were made between RNA extracted from samples snap frozen in liquid nitrogen and those treated with an RNA preservative before freezing. Further comparisons were made between biopsies taken by different methods. Acceptable RNA for gene expression analysis was extracted from 72% of lung cancer biopsies. Use of an RNA preservative for storage allowed the extraction of higher quality, more intact RNA from biopsies gathered by both endobronchial forceps and transbronchial needle aspiration. High quality RNA could also be extracted from computed tomography-guided needle core biopsies. Banking lung cancer biopsy specimens by storage in an RNA preservative solution will allow use of a broader spectrum of lung cancers for gene expression analysis. We describe a model that makes personalized medicine for lung cancer patients a more practical proposition.

Molecular changes in epidermal growth factor receptor (EGFR) in non-small cell lung cancer (NSCLC) biopsies at time of progression compared to initial biopsy

e22066 Background: EGFR mutations predict sensitivity and clinical outcome to tyrosine kinase inhibitors (TKI) in NSCLC. The two most commonly described mutations are Exon 19 deletion and Exon 21 L858R missense mutations. Genetic alterations over time have been described in other tumour types, but studies assessing EGFR genotypic changes with lung cancer progression are lacking. We sought to compare EGFR mutational status from lung tumors at time of recurrence or progression with the primary tumor. Methods: Using the Jewish General Hospital lung cancer database, of all patients diagnosed with NSCLC since 1999, those with biopsies at two different points in time were identified. All tumour samples were genotyped for EGFR exons 19 and 21 mutations using denaturing high performance liquid chromatography (dHPLC). Results: 29 patients were identified. Data for 12 patients, whose time of recurrence or progression varied between 4 months and 6 years, are available at this time. Of 12 patients, one had EGFR exon 19 mutation at time of diagnosis. One patient who initially displayed no EGFR mutation was found to have an exon 19 deletion at time of recurrence. The one with exon 19 at time of initial diagnosis continued to express exon 19 in the second biopsy. Conclusions: To our knowledge, this is the only study assessing changes in molecular genotype using dHPLC between primary and recurrent or progressive lung cancer biopsy specimens. Although sample size is small, it is evident that changes in EGFR mutational status can occur. Further prospective studies are required to determine how commonly molecular changes occur. No significant financial relationships to disclose.

Molecular signatures in biopsy specimens of lung cancer.

Gene expression profiles of resected tumors may predict treatment response and outcome. We hypothesized that profiles derived from lung tumor biopsies would discriminate tumor-specific gene signatures and provide predictive information about outcome. Lung carcinoma specimens were obtained from 23 patients undergoing computed tomography-guided transthoracic biopsy or endobronchial brushing for undiagnosed nodules. Excess tissue was processed for gene profiling. We built class prediction models for lung cancer histology and for cancer outcome. The histology model used an F test to identify 99 genes that were differentially expressed among lung cancer subtypes. The histology validation set class prediction accuracy rate was 86%. The outcome model used the maximum difference subset algorithm to identify 42 genes associated with high risk for cancer death. The outcome training set class prediction accuracy rate was 87%. In conclusion, gene expression profiles of biopsy specimens of lung cancers identify unique tumoral signatures that provide information about tissue morphology and prognosis. The use of specimens acquired from lung biopsy procedures to identify biomarkers of clinical outcome may have application in the management of patients with lung cancer. The procedures are safe and feasible; the efficacy and utility of this strategy will ultimately be determined by prospective clinical trials.

Enhancement of Apo2L/TRAIL (tumor necrosis factor-related apoptosis-inducing ligand)-induced apoptosis in non-small cell lung cancer cell lines by chemotherapeutic agents without correlation to the expression level of cellular protease caspase-8 inhibitory protein.

Apo2L/tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a potential anticancer drug that promotes apoptosis specifically in tumor cells. Because not all cancer cells are susceptible to Apo2L/TRAIL, the aim of our study was to determine whether non-small cell lung cancer cells can be sensitized by chemotherapeutic agents for Apo2L/TRAIL-induced apoptosis. In addition, endogenous expression levels of the caspase-inhibiting cellular protease caspase-8 inhibitory protein (C-FLIP) were measured to investigate partial resistance to Apo2L/TRAIL. Six human lung cancer cell lines (A549, NCI-H358, Calu1, Calu6, SkMes1, and SkLu1) were incubated with soluble Apo2L/TRAIL and two different concentrations each of cisplatin, paclitaxel, doxorubicin, 5-fluorouracil, and camptothecin. After 24 hours the rate of apoptosis was measured by annexin V/propidium iodide staining followed by FACScan analysis. Expression levels of C-FLIP in cell lines and lung cancer biopsy specimens were determined by Western blotting. Treatment of lung cancer cells with Apo2L/TRAIL alone resulted in apoptotic cell death in four cell lines (P <.001). Combining Apo2L/TRAIL and chemotherapeutic agents enhanced the rate of apoptosis significantly. Statistical analysis revealed a synergistic effect of Apo2L/TRAIL in combination with 1.8 mmol/L camptothecin and 100 micromol/L cisplatin, each in four of the six cell lines (P <.002). Western blot analysis showed that sensitization to Apo2L/TRAIL did not correlate with the expression of cellular protease caspase-8 inhibitory protein. Furthermore, no increased cellular protease caspase-8 inhibitory protein levels relative to those in normal lung tissue could be found in non-small cell lung cancer specimens from 12 patients. Apo2L/TRAIL-induced apoptosis in non-small cell lung cancer cell lines is significantly enhanced by chemotherapeutic agents. Resistance and sensitization to Apo2L/TRAIL are not correlated with the endogenous expression level of cellular protease caspase-8 inhibitory protein, implying that in non-small cell lung cancer other mechanisms are responsible for inhibition of the Apo2L/TRAIL pathway. Even though the molecular mechanism remains unclear, the combination of Apo2L/TRAIL with chemotherapy may be a promising treatment modality for non-small cell lung cancer.

A comparative study of telomerase activity in sputum, bronchial washing and biopsy specimens of lung cancer

The potential of telomerase, the ribonucleoprotein enzyme, as a non-invasive screening marker was studied in pre-bronchoscopy sputum (S), bronchial washings (W) and bronchoscopic biopsy (B) samples from individuals under evaluation for lung cancer. Out of the 52 cases studied, 42 were clinically suspected primary lung cancer patients and 10 had pulmonary disorders but had no clinical evidence of lung cancer. Fifteen (39.5%) S samples, 24 (63.1%) W samples and 32 (84.2%) B samples, which were cytologically/histopathologically positive were also positive for telomerase activity. Interestingly, 16 (42%) S samples, 20 (52.6%) W samples and 20 (52.6%) B samples, initially reported cytologically/histopathologically negative, showed detectable telomerase activity. Lung cancer was finally confirmed in these cases by repeat cytology/histopathology. However, telomerase activity was detected in 31 (81.6%) S, 26 (68.4%) W and 33 (86.8%) B samples of suspected lung cancer patients. Telomerase activity was negative in S, W, and B of four of the suspected cases, which ultimately turned out to be negative for lung cancer. Cytopathology/histopathology alone (including repeat attempts) identified 15 (39.5%) cases of sputum, 24 (63.1%) cases of bronchial washings and 32 (84.2%) bronchoscopic biopsy samples. Out of 10 controls, low telomerase activity was detected in only one (10%) of the bronchial washings, which later turned out to be due to large number of inflammatory cells. Telomerase activity assay of sputum carried sensitivity, specificity and diagnostic accuracy of 81.6, 100 and 86.5%, respectively, while that for bronchial washing was 68.4, 100 and 76.9%, respectively, and for bronchoscopic biopsy samples was 86.8, 100 and 88.1%, respectively. A positive correlation ( P<0.01) was seen between age and telomerase activity in sputum, bronchial washing and biopsy samples but no significant correlation was seen between sex and telomerase activity or duration of smoking and telomerase activity. A significant positive correlation was observed between staging and telomerase activity in sputum ( P<0.01), bronchial washing ( P<0.01) and biopsy samples ( P<0.01). Our findings indicate that telomerase is a specific marker for malignant lung disease and can complement cytology/histopathology in the diagnosis of lung cancer. Sputum telomerase assay holds the potential for early and non-invasive diagnosis of lung cancer.

Expression of IGF-1, TGF-α and p53 protein in imprint smears of bronchial biopsy specimens of lung cancer: Orphanidou D, Athanasiadou P, Toumbis M, Rasidakis A, Petrakakou E, Dimakou K et al. Department of Respiratory Medicine, Medical School University, Chest Diseases Hospital, 152 Mesogion Ave., Athens 115 27. Oncol Rep 1996; 3: 75–80

Expression of IGF-1, TGF-α and p53 protein in imprint smears of bronchial biopsy specimens of lung cancer: Orphanidou D, Athanasiadou P, Toumbis M, Rasidakis A, Petrakakou E, Dimakou K et al. Department of Respiratory Medicine, Medical School University, Chest Diseases Hospital, 152 Mesogion Ave., Athens 115 27. Oncol Rep 1996; 3: 75–80

Expression of IGF-I, TGF-alpha and p53 protein in imprint smears of bronchial biopsy specimens of lung cancer

The expression of insulin-like growth factor-I (IGF-I), transforming growth factor-alpha (TGF-alpha) and p53 protein was examined in bronchial biopsy imprint smears of patients with primary lung cancer and benign lung disorders, by immunohistochemistry. Of the 44 malignant imprint smears, 26 (59%) were positively stained for IGF-I, 18 (41%) for TGF-alpha and 29 (66%) for p53 protein. In contrast, of the 36 benign imprint smears none was positively stained for IGF-I (p<0.001), whereas 7 were positively stained for TGF-alpha (p>0.05) and 3 for p53 protein (p<0.001). There was no correlation between the expression of the examined markers and the histological type of lung cancer. The most sensitive indicator of malignancy was p53 overexpression (65.9%), the most specific was IGF-I (100%) whereas both revealed 77.5% accuracy. The combination of IGF-I and p53 revealed 75% sensitivity, 91.6% specificity and 82.5% accuracy. When one marker was positive the relative possibility of lung cancer was 67.1%. This possibility increased to 77.7% when two markers were positive and to 100% when three markers were positive. These results suggest that the evaluation of IGF-I and p53 in imprint smears could be of value in diagnosis of lung cancer.