Bronchoscopy in Determining the Effect of E-Cigarette Smoking on Biomarkers in the Lungs

Obtained via bronchoscopy with saline bronchoalveolar lavage (BAL) and bronchial brushings. Descriptive statistics and clustering (e.g., principal components analysis [PCA]), will be performed for all biomarker data measured at the different visits. Baseline data (1st bronchoscopy) will then be compared between the four groups (3 conditions and control) using a one-way analysis of variance (ANOVA). The non-parametric Kruskal-Wallis test (continuous or ordinal variables) or the chi-square test (categorical variables) will be applied when the normality assumption of the data is not met. Controls will be compared to complete substitution with the standardized research electronic cigarettes (SREC) (with and without nicotine) or nicotine replacement therapy (NRT. Generalized linear models (GLM) will be employed with measurement (cell count, gene expression, etc.) as the dependent variable, a covariable for baseline measure, and a main effect of arm.

Obtained via bronchoscopy with saline bronchoalveolar lavage (BAL) and bronchial brushings. Descriptive statistics and clustering (e.g., principal components analysis [PCA]), will be performed for all biomarker data measured at the different visits. Baseline data (1st bronchoscopy) will then be compared between the four groups (3 conditions and control) using a one-way analysis of variance (ANOVA). The non-parametric Kruskal-Wallis test (continuous or ordinal variables) or the chi-square test (categorical variables) will be applied when the normality assumption of the data is not met. Controls will be compared to complete substitution with the standardized research electronic cigarettes (SREC) (with and without nicotine) or nicotine replacement therapy (NRT. Generalized linear models (GLM) will be employed with measurement (cell count, gene expression, etc.) as the dependent variable, a covariable for baseline measure, and a main effect of arm.

FeNO will be measured with the NIOX VERO. Descriptive statistics and clustering (e.g. PCA), will be performed for all biomarker data measured at the different visits. Baseline data (1st bronchoscopy) will then be compared between the four groups (3 conditions and control) using a one-way ANOVA. The non-parametric Kruskal-Wallis test (continuous or ordinal variables) or the chi-square test (categorical variables) will be applied when the normality assumption of the data is not met. Controls will be compared to complete substitution with the SREC (with and without nicotine) or NRT. GLM will be employed with measurement (cell count, gene expression, etc.) as the dependent variable, a covariable for baseline measure, and a main effect of arm.

Total RNA containing small RNAs will be extracted from bronchial brushing specimens using commercially available kits and used for both gene expression using the Affymetrix GeneChip Human Transcriptome Array and for miRNA expression using the Affymetrix GeneChip miRNA Array. Expression will separately be assessed through RNA sequencing. Total RNA will be extracted from lavage cells and saliva for comprehensive profiling of microbiome using commercially available kits. The ?omics? (miRNA, mRNA, and metabolomics) analysis and visualization of the data will be performed in the R statistical language. Data will be log2-transformed and normalized using either quantile normalization for gene expression (miRNA and mRNA) or normalization for metabolomics. Unsupervised clustering analysis, including PCA and hierarchical clustering will be performed to visualize natural clusters in the dataset and evaluate data quality.

Total DNA will be extracted using commercially available kits and used for genome-wide DNA methylation profiling. Total DNA including mitochondrial DNA (mtDNA) will be extracted using commercially available kits and used for mtDNA mutation using Hiseq Next Generation Sequencing (NGS) and for mtDNA contents using quantitative polymerase chain reaction (qPCR). For preliminary identification of patterns in DNA methylation, unsupervised hierarchical clustering among the groups of samples will be performed. The Euclidian distance among the groups of samples will be calculated by the average linkage. In order to assess variance among samples, PCA will be done.

Obtained via bronchoscopy with saline BAL and bronchial brushings. Samples will be analyzed to determine their bacterial composition.

Obtained via bronchoscopy with saline BAL and bronchial brushings. The omics (miRNA, mRNA, and metabolomics) analysis and visualization of the data will be performed in the R statistical language. Data will be log2-transformed and normalized using either quantile normalization for gene expression (miRNA and mRNA) or normalization for metabolomics. Unsupervised clustering analysis, including PCA and hierarchical clustering will be performed to visualize natural clusters in the dataset and evaluate data quality.