Early Detection of Lung Cancer by Molecular Markers in Endobronchial Lining Fluid

Early Detection of Lung Cancer by Molecular Markers in Endobronchial Epithelial Lining Fluid

Nicolas Kahn, Michael Meister, Ralf Eberhardt, Thomas Muley, Philipp A. Schnabel, Christian Bender, Marc Johannes, Denise Keitel, Holger Sültmann, Felix JF. Herth, Ruprecht Kuner

Online Data Supplement

1. Methods

2. Supplementary Table 1 - 5

Methods

Patient Cohort

This study included 71 patients with indeterminate solitary pulmonary nodules (IPN) who were scheduled to undergo diagnostic bronchoscopy at the Thoraxklinik Heidelberg between 2007 and 2010. The Institutional Review Board approved the data collection and analysis, and written informed consent was obtained from all patients prior to the procedure. Definitive histological diagnosis was established either by transbronchial biopsy or by subsequent surgical resection, and included adenocarcinoma (AC, n=31), squamous cell carcinoma (SCC, n=14), large cell lung carcinoma (LCLC, n=4), mixed type (n=1), carcinoid (n=1), benign nodule (n=20). Patient characteristics and ELF sampling parameters are given in Table 1. Wilcoxon, Fisher’s exact and Chi-square tests were used to compare parameters associated with the diagnosis of nodules. Patients were separated into a screening (n=15) and a validation cohort (n=56) (Figure 1). Detailed description of each patient and the experimental assignment is listed in

Table E1.

Bronchoscopic Microsampling

In summary, the procedure was performed undergeneral anesthesia or conscious sedation in a standardized fashion. A standard BAL catheter was placed through the working channel in peripheral lung and sampling of endobronchial lining fluid was performed by using a specific mucus collection probe (Olympus, BC-401C/BC402C). The bronchoscopic microsampling device (BMS) was inserted through the guide catheter into the bronchi, positioned fluoroscopically, and placed at the nearest possible point next to the lesion. The collection probe containing the absorbed ELF was sectioned and immediately shock-frozen in liquid nitrogen. In each patient, BMS procedures were repeated twice at the site of the lesion (IPN) and at the corresponding subsegmental bronchus of the contralateral lung (CL) as internal control.

RNA extraction and microarray experiment

Total RNA extraction of ELF samples was performed using RNeasy Micro Kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. The quantity of total RNA was measured by spectrophotometer and the quality was assessed by capillary electrophoresis (Agilent Technologies GmbH, Waldbronn, Germany). For microarray experiments, we selected total RNA samples from nodules and contralateral site, which yielded higher concentration and quality (total amount of RNA > 200ng, RIN value > 5).

We used the microarray platform Affymetrix GeneChip U133 Plus 2.0 (Affymetrix, Santa Clara, CA, US) covering about 47,000 transcripts and variants. The microarray experiment comprised 30 ELF samples of matched nodule and contralateral site (15 individuals). Total RNA (150 ng) was amplified using GeneChip® 3’ IVT Express Kit (Affymetrix). Labeling, hybridization and scanning was processed according to the manufacturers’ instructions (Affymetrix).

The microarray data were normalized by the gcRMA method as previously described (1), and deposited into the NCBI GEO database (GSE27489). Linear models for microarray data software package (LIMMA) (2) was used to identify differentially expressed genes between ELF samples of malignant (or AC and SCC respectively) and benign nodules dependent (ratio IPN/ CL) or independent (IPN only) of the contralateral site. The genes were ranked according to the LIMMA score and by the linear fold change.

Quantitative real-time PCR

In the validation study, qRT-PCR was used to analyze gene expression in 112 independent specimens including matched indeterminate pulmonary nodule and contralateral ELF extracts of the same patients (n = 56). Total RNA was reversely transcribed including oligo d(T) and random primers (50 pmol each) with the RevertAid™ First Strand cDNA Synthesis Kit (Fermentas, Burlington, ON, Canada). Expression analysis of 16 genes was done for each sample in technical duplicates by using the quantitative real-time PCR system (ABI Prism 7900HT Sequence Detection System; Applied Biosystems, Weiterstadt, Germany) and gene specific primers and probes (Taqman assays, ABI). Relative quantification was done using two internal control genes esterase D (ESD) and polymerases 2A (POLR2A) as described (1). A detailed list of examined genes is given in Table E2. The qRT-PCR data were deposited into the NCBI GEO database (GSE27553). Statistical analysis was done using the paired or unpaired Wilcoxon test, respectively. Genes were regarded to be differentially expressed with a p-value < 0.05 and a fold change < 0.5 or > 2. We accounted for the interaction between biomarker expression and significant clinical parameters (nodule diameter, age) by using ANOVA analysis. The discriminative efficacy of various parameters (nodule diameter, TNC expression and multiplying both values) was analysed by receiver operating characteristics (ROC) curves using the area under the curve (AUC) as a measure. All p-values were derived from 2-tailed statistical tests, and a value of 0.05 or less was regarded as statistically significant.

The microarray and the qRT-PCR data were deposited MIAME-compliant into the NCBI GEO database as super series (GSE27554).

References

E1Kuner R, Muley T, Meister M, Ruschhaupt M, Buness A, Xu EC, Schnabel P, Warth A, Poustka A, Sültmann H, Hoffmann H. Global gene expression analysis reveals specific patterns of cell junctions in non-small cell lung cancer subtypes. Lung Cancer 2009;63:32-38.

E2Smyth GK. Linear models and empirical Bayes methods for assessment differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology 3, 2004;No. 1, Article 3.