US20210254171A1

US20210254171A1 - Gene expression-based biomarker for the detection and monitoring of bronchial premalignant lesions

Info

Publication number: US20210254171A1
Application number: US17/182,044
Authority: US
Inventors: Jennifer E. Beane-Ebel; Anna Tassinari; Avrum Spira; Marc E. Lenburg
Original assignee: Boston University
Current assignee: Boston University
Priority date: 2016-07-08
Filing date: 2021-02-22
Publication date: 2021-08-19
Also published as: US10927417B2; WO2018009915A1; US20180010197A1

Abstract

Disclosed herein are assays and methods for the identification of premalignant lesions, as well as methods of determining the likelihood that such premalignant lesions will progress to lung cancer. Also disclosed are methods and assays that are useful for monitoring the progression of premalignant lesions to lung cancer. The assays and methods disclosed herein provide minimally invasive means of accurately detecting and monitoring the presence or absence of premalignant lesions, thus providing novel insights into the earliest stages of lung cancer and facilitating early detection and early intervention.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. application Ser. No. 15/644,721, filed Jul. 7, 2017, which claims the benefit of U.S. Provisional Application No. 62/360,218, filed on Jul. 8, 2016, the contents of which are hereby incorporated by reference in its entirety.

BACKGROUND OF THE INVENTION

Lung cancer (LC) is the leading cause of cancer death in the United States. The molecular events preceding the onset of LC and the progression of premalignant lesions (PMLs) to lung cancer are poorly understood. This is due in part to the lack of reliable biomarkers which complicates the study of such lesions. Currently there are no molecular tests to identify PMLs or describe their changes over time. The only technology that is able to visualize and sample premalignant lesions is auto-fluorescent bronchoscopy, which is limited in sensitivity and is not in widespread clinical use.
Needed are novel biomarkers, methods and assays that are capable of facilitating the evaluation of PMLs. Suspicious lesions on chest computed tomography (CT) scans typically prompt bronchoscopic evaluation, which is also limited by varying diagnostic yields. Moreover, negative bronchoscopies prove a clinical dilemma, whereby the need to provide a diagnostic answer is countered by the invasiveness of follow-up studies.
A previously reported biomarker, PERCEPTA® (Veracyte Inc.), has demonstrated the potential benefit of employing a bronchial gene expression-based classifier on a sub-set of patients with non-diagnostic bronchoscopies, through modifying risk stratification of patients. However, this biomarker has demonstrated greatest benefit amongst those with a moderate pre-test probability with modest overall sensitivities. The employment of a novel pre-malignancy marker would complement the PERCEPTA® biomarker in this sub-set of patients, facilitating the identification of those patients that would be at high risk for PML progression.
Also needed are new biomarkers, methods and assays for use in lung cancer screening assays and the early detection of PMLs. A recent large randomized controlled trial has led to the recent endorsement of annual lung cancer screening with low dose CT for asymptomatic patients that are at higher lung cancer risk. This has created a large volume of chest CTs, whose performance is marred by the high rate of false positive results. It is anticipated that this will lead to a large need for invasive procedures for benign disease. A pre-malignancy biomarker could complement the diagnostic work up of lesions identified through screening, which are typically more complicated since such lesions identified on screening are usually smaller and more complex. Additionally, patient screening eligibility is based solely on epidemiological and demographic considerations, which still vary between different proposed guidelines. This leads to varying referral patterns and missed opportunities to screen a large proportion of those patients with high risk that do not meet dictated criteria. The availability of biomarkers, methods and assays for the detection of PMLs would overcome this challenge by facilitating the identification of pre-malignancy-associated changes and risk of progression, would provide a first step to identifying molecular risk factors for lung cancer, and would identify those patients who would benefit from CT screening. Such biomarkers would also be useful for patent risk stratification, which would assist in the identification of those patients that may benefit from additional screening of those patients harboring premalignant molecular alterations, which could in turn inform future decision making.
The limited understanding of the mechanisms involved in transforming PMLs into LC has restricted the ability to intervene in these processes, making the identification of chemoprevention agents difficult in view of the challenges involved in discerning premalignant phenotypes through currently available means. Furthermore, clinical trials in this space are exceedingly difficult given the long duration required to detect significant outcome benefits. Accordingly, biomarkers, assays and methods that are reflective of pre-malignancy would facilitate “smart” patient enrollment for trials and would allow accounting for molecular heterogeneity involved in random patient recruitment in such trials.

SUMMARY OF THE INVENTION

The present inventions provide insight into the mechanisms that are involved in the transformation or progression of premalignant bronchial lesions into lung cancer. Provided herein are novel biomarkers, methods and assays that are useful in lung cancer screening and the early detection of premalignant lesions (PMLs). The biomarkers, methods and assays of the present invention also facilitate the monitoring of PMLs and their progression or regression over time. Advantageously, the assays and methods disclosed herein may be rapidly performed in a non-invasive or minimally-invasive manner, providing objective results, contributing to the identification and monitoring of subjects that are suspected of having PMLs, facilitating the clinical decision making of the treatment of such subjects and informing clinical trial recruitment efforts.
In certain aspects, the biomarkers, methods and assays disclosed herein may be assessed or performed on a biological sample that is obtained from a subject at a site that is distal to the suspected site of the premalignant bronchial lesion. For example, in certain embodiments, the assays and methods of determining the presence of PMLs or cancer in the lungs may be performed by determining the expression of one or more genes in nasal or buccal epithelial cells and/or tissues. Similarly, such assays and methods may be performed by determining the expression of one or more genes in the subject's peripheral blood cells. In certain aspects, the biomarkers, methods and assays disclosed herein may be assessed or performed on, or additionally include, a biological sample that is obtained from a subject with a positive result in an imaging study (e.g., chest X-ray, CT scan, etc.). In some aspects, the methods and assays disclosed herein can comprise a step of performing an imaging study. In certain aspects, the biomarkers, methods and assays disclosed herein may be assessed or performed on, or additionally include, a biological sample that is obtained from a subject with a positive result in an imaging study (e.g., chest X-ray, CT scan, etc.) to confirm or rule out the positive result. In some aspects, the methods or assays disclosed herein are used to determine whether a positive result in an imaging study warrants a further invasive procedure (e.g., bronchoscopy), chemoprophylaxis, and/or chemotherapy.
In some embodiments, methods and assays disclosed herein may be assessed or performed on a biological sample that is obtained from a subject at a suspected site of a PML (e.g., premalignant bronchial lesion). In some embodiments, the suspected site is identified as having abnormal fluorescent during auto-fluorescence bronchoscopy, although the method of identifying the suspected site is not limited. In some embodiments, the methods and assays disclosed herein may be performed on a biopsy of a suspected PML as an alternative to, or in addition to, a histological examination of the biopsy.
In certain aspects, disclosed herein are methods of determining the presence or absence of a premalignant lesion in a subject. Such methods comprise the steps of: (a) measuring a biological sample comprising airway epithelial cells of the subject for expression of one or more genes; and (b) comparing the expression of the one or more genes to a control sample of those genes from individuals without premalignant lesions; wherein the one or more genes are selected from the group consisting of genes in Table 3, and wherein differential expression of the subject's one or more genes relative to the control sample is indicative of the presence of a premalignant lesion in the subject. Similarly, in certain embodiments, non-differential expression of the subject's one or more genes relative to the control sample is indicative of the absence of a premalignant lesion in the subject.
Also disclosed herein are methods of determining the likelihood that a premalignant lesion in a subject will progress to lung cancer. In certain aspects, such methods comprise the steps of: (a) measuring a biological sample comprising airway epithelial cells of the subject for expression of one or more genes; and (b) comparing the expression of the one or more genes to a control sample of those genes from individuals with lung cancer; wherein the one or more genes are selected from the group consisting of genes in Table 3, and wherein differential expression of the subject's one or more genes relative to the control sample is indicative of a low likelihood of the premalignant lesion progressing to lung cancer. In some embodiments, non-differential expression of the subject's one or more genes relative to the control sample is indicative of a high likelihood of the premalignant lesion progressing to lung cancer.
In certain embodiments, also disclosed herein are methods of monitoring whether a premalignant lesion will progress to lung cancer in a subject. Such methods comprise subjecting a biological sample comprising airway epithelial cells of the subject to a gene expression analysis, wherein the gene expression analysis comprises comparing gene expression levels of one or more genes selected from the group of genes in Table 3 to the expression levels of a control sample of those genes from individuals with cancer, and wherein differential expression of the subject's one or more genes relative to the control sample is indicative of a lack of progression of the premalignant lesion to lung cancer. Similarly, in certain aspects non-differential expression of the subject's one or more genes relative to the control sample is indicative of progression of the premalignant lesion to lung cancer.
In yet other embodiments, also disclosed herein are methods of determining the presence of a premalignant lesion in a subject comprising the steps of: (a) measuring a biological sample comprising airway epithelial cells of the subject for expression of one or more genes; and (b) comparing the expression of the one or more genes to a control sample of those genes obtained from individuals without premalignant lesions; wherein the one or more genes are selected from the group of genes in at least one pathway in Dataset 2, and wherein differential expression of the subject's one or more genes relative to the control sample is indicative of the presence of a premalignant lesion in the subject. In some embodiments, non-differential expression of the subject's one or more genes relative to the control sample is indicative of the absence of a premalignant lesion in the subject.
In certain aspects of any of the foregoing methods, at least two genes, at least five genes, at least ten genes, at least twenty genes, at least thirty genes, at least forty genes, at least fifty genes, at least one hundred genes, at least two hundred genes or at least two hundred and eighty genes are measured. In some embodiments of the foregoing methods, the one or more genes comprise those genes associated with a pathway identified in Dataset 2.
In some embodiments of any of the foregoing methods the airway epithelial cells comprise bronchial epithelial cells. In certain aspects, such bronchial epithelial cells are obtained by brushing the bronchi walls of the subject. In certain aspects of any of the foregoing methods, the airway epithelial cells comprise nasal epithelial cells. In certain aspects of any of the foregoing methods, the airway epithelial cells comprise buccal epithelial cells. In still other embodiments of the present inventions, the airway epithelial cells do not comprise bronchial epithelial cells. In some embodiments, the airway epithelial cells are obtained from a suspected PML site (e.g., abnormal fluorescing areas during auto-fluorescence bronchoscopy).
In certain aspects, the methods disclosed herein are performed with, or further comprise assessing or determining one or more of the subject's secondary factors that affect the subject's risk for having or developing lung cancer. For example, in some embodiments, one or more secondary factors are selected from the group consisting of advanced age, smoking status, the presence of a lung nodule greater than 3 cm on CT scan and time since quitting smoking. In certain embodiments of the foregoing methods, expression of the one or more genes is determined using a quantitative reverse transcription polymerase chain reaction, a bead-based nucleic acid detection assay or an oligonucleotide array assay.
The foregoing methods are useful for predicting or monitoring the progression of PMLs to lung cancer. For example, a lung cancer selected from the group consisting of adenocarcinoma, squamous cell carcinoma, small cell cancer or non-small cell cancer.
In some embodiments, the one or more genes comprise mRNA and/or microRNA. In some embodiments, the differential expression is determined by reverse transcribing one or more RNAs of the one or more genes into cDNA in vitro. In some aspects, the one or more genes comprise cDNA. In yet other embodiments, the one or more genes are labeled prior to the measuring.
The above discussed, and many other features and attendant advantages of the present inventions will become better understood by reference to the following detailed description of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawings will be provided by the Office upon request and payment of the necessary fee.

FIG. 1 represents a flow diagram depicting the design of the study used in the Examples. Depicted is the use of bronchial brushings collected from subjects with (red, n=50) and without (gray, n=25) PMLs from the BCCA as part of the BC-LHS for differential gene expression/pathway analysis and for biomarker development. Independent human and mouse bronchial biopsies and biopsy cell cultures were used to validate these findings via mitochondrial enumeration, bioenergetics, and immunohistochemistry (left panel). Biomarker development was conducted by splitting samples from the BC-LHS into a discovery (n=58) and a validation set (Validation 1, n=17) (right panel). The discovery set was used to create the gene expression-based biomarker to detect the presence of PMLs in the airway field of injury. The biomarker was tested on the BC-LHS validation set and an external validation set (bottom) from RPCI (Validation 2, n=28 matched time point pairs, stable/progressing pairs in yellow and regressing pairs in blue).

FIG. 2 shows an unsupervised hierarchal clustering of genes associated with the presence of premalignant lesions. Residual gene expression of the 280 genes differentially expressed between subjects with PMLs (red) and without PMLs (gray). Top color bars represent the worst biopsy histological grade observed during bronchoscopy and genomically-derived smoking status of the subjects. The 14 genes in the KEGG oxidative phosphorylation pathway are indicated in cyan. The residual values after adjusting for the 7 surrogate variables were z-score normalized prior to Ward hierarchal clustering.

FIGS. 3A-3E illustrate OXPHOS up-regulation in premalignant lesion biopsies. FIG. 3A shows the mean baseline OCR/ECAR ratio measured in human bronchial biopsies cultures from PMLs (pink, n=6) was 2.5 fold higher than the biopsies of normal airway epithelium (gray n=6) (p=0.035). Error bars represent standard error of the mean. FIG. 3B shows bioenergetic studies testing mitochondrial function demonstrate PMLs (pink) have a significantly (˜1.5 fold) higher maximal respiration (p=0.022). Error bars represent standard error of the mean. FIG. 3C and FIG. 3D show mitochondrial enumeration by FACS analysis of MitoTraker GFP suggests increased OCR is not reliant on increase mitochondria as the difference in GFP per cell was not significant (p=0.150). FIG. 3E shows representative images of TOMM22 and COX IV staining in which expression of both proteins is increased in low and moderate dysplastic lesions in both human and NTCU-mouse PMLs. (Magnification 400×).

FIGS. 4A-4C shows that PML-associated gene expression alterations in the field are concordant with SCC-related datasets. The genes up-regulated in the field of subjects with PMLs are red and genes down regulated in blue. GSEA identified the significant enrichment of the lung cancer-related gene expression signatures shown in this ranked list. The black vertical lines represent the position of the genes in the gene set in the ranked list and the height corresponds to the magnitude of the running enrichment score from GSEA. FIG. 4A shows top differentially expressed genes from analysis of TCGA RNA-Seq data comparing lung SCC and matched adjacent normal tumor tissue. FIG. 4B shows Ooi et al. gene sets for early gene expression changes defined by genes altered between premalignant and normal tissue and between tumor and normal tissue (p<0.05) using laser capture microdissected (LCM) epithelium from the margins of resected SCC tumors. FIG. 4C shows top differentially expressed genes from analysis of cytologically normal bronchial epithelial cells from smokers with and without lung cancer (GSE4115).

FIGS. 5A-5B show performance of an airway biomarker in detecting the presence and progression of premalignant lesions. The ROC curves demonstrate the biomarker performance. FIG. 5A is a ROC curve (AUC=0.92) showing biomarker performance based on predictions of the presence of PMLs in the validation samples (n=17), red line. Shuffling of class labels (n=100 permutations) produced an average ROC curve (black line) with a significantly lower AUC (p<<0.001). FIG. 5B is a ROC curve (AUC=0.75) showing biomarker performance based on changes in biomarker score over time in detecting PML regression or stable/progression.

FIG. 6 shows unsupervised hierarchal clustering of genes associated with smoking status. The weighted voting algorithm was trained on z-score normalized microarray data (GSE7895) across 94 genes differentially expressed between current and never smokers and used to predict smoking status in log 2-transformed counts per million (cpm) that were z-score normalized from the 82 mRNA-Seq samples. The heatmap shows the results of unsupervised Ward hierarchal clustering across the 82 mRNA-Seq samples and the 94 genes. The row color label indicates if genes were up-regulated (red) or down-regulated (green) in current smokers compared to never smokers in GSE7895. The lower column color labels indicate the smoking status in the clinical annotation (self-report) with light gray indicating former smokers and dark gray indicating current smokers. The upper column color labels indicate the predicted class of the samples based on the 94 genes with white indicating former smokers and black indicating current smokers. Log 2-cpm mRNA-Seq data was z-score normalized prior to clustering.

FIGS. 7A-7H show cellular metabolism in cancer cell lines and in the airway field associated with premalignant lesions FIG. 7A shows GSVA scores were calculated based on genes in KEGG OXPHOS pathway and KEGG, Biocarta, and Reactome Glycolysis pathways in the CCLE cell lines highlighting the H1229 (green) (high OXPHOS and moderate glycolysis), SW900 (red) (moderate OXPHOS and low glycolysis) and H2805 (blue) ((low OXPHOS and moderate glycolysis). FIG. 7B shows baseline OCR/ECAR ratio values for the cancer cells lines demonstrating the relationship between elevated OXPHOS GSVA scores and oxygen consumption. FIG. 7C shows elevation of respiratory capacity associated with high OXPHOS gene score in response to mitochondrial perturbation. FIG. 7D shows elevated ECAR response in the H1299 and H205 is associated with the moderate glycolysis GSVA score, however, although the SW900 glycolysis GSVA scores agree with baseline ECAR, in the state of repressed OXPHOS, glycolysis is activated. FIG. 7E shows enumeration of mitochondria within each cancer cell suggests that increased GSVA scores for OXPHOS or glycolysis did not correlate with mitochondrial number. H2085 cells had the lowest OXPHOS GSVA score, the lowest basal OCR, and the lowest respiratory capacity, but their mitochondrial content was significantly greater that H1299 and SW900 (p=0.03). FIG. 7F shows cell area (FSC-A) is correlated with mitochondrial number (fluorescence of MitoTracker Green FM). FIG. 7G shows GSVA scores were calculated based on genes in KEGG OXPHOS pathway. The GSVA scores for OXPHOS activity were significantly elevated in the airway field of subjects with PMLs compared to subjects without PMLs (p<0.01). FIG. 7H shows GSVA scores were calculated based on genes in the KEGG, Biocarta, and Reactome Glycolysis pathways. The mean GSVA scores were moderately elevated in the airway field of subjects with PMLs compared to subjects without PMLs.

FIG. 8 shows a biomarker discovery flowchart. Samples (n=75) were split into a discovery set (n=58) and a validation set (n=17). The pipeline was run 500 times, and each time the discovery set was randomly split into training (80% of samples, n=46) and test (20% of samples, n=12) sets. The training set samples were used to train the biomarker using all combinations of pipeline parameters, including: 1. Up-/down-regulation ratio: TRUE or FALSE; 2. Data type: raw counts, RPKM or CPM; 3. Gene filter: genes with signal in at least 1%, 5%, 10%, or 15% of samples; 4. Feature selection: edgeR, edgeR correcting for gb-ratio, limma, limma correcting for gb-ratio, glmnet, random forest, DESeq, SVA, or partial AUC; 5. Gene number: 10, 20, 40, 60, 80, 100, or 200 genes (see Biomarker size); and 6. Prediction method: weighted voting, random forest, SVM, naïve bayes, or glmnet.

FIG. 9 shows that biomarker predicts dysplasia status in bronchial biopsies. ROC curve demonstrates the performance of the biomarker in distinguishing between premalignant lesion biopsies (severe=8, moderate=25, and mild dysplasia=14) and biopsies with normal histology (normal=24 and hyperplasia=20). Biomarker achieved AUC of 72% (with a 62%-83% confidence interval), sensitivity of 81% (38 of 47 dysplastic biopsies predicted correctly), and specificity of 66% (29 of 44 normal biopsies predicted correctly).

DETAILED DESCRIPTION OF THE INVENTION

Lung cancer develops in a sequenced manner Patches of lung cells gain the ability to multiply faster than their neighboring normal cells by acquiring mutations and these patches of cells are called “premalignant lesions” or “PMLs.” Some of these PMLs may progress to lung cancer. The inventions disclosed herein are based upon a biomarker that is capable of identifying and distinguishing epithelial cells from a person with lung cancer from normal epithelial cells. In particular, the inventions disclosed herein are based on the findings that exposure to carcinogens such as cigarette smoke induces smoking-related mRNA and microRNA expression alterations in the cytologically normal epithelium that lines the respiratory tract, creating an airway field of injury (1-8). Such gene expression alterations that were observed in the airway field of injury were used to develop a diagnostic test to facilitate early lung cancer lung cancer detection (9-12). Examination of gene signatures for p63 and the phosphatidylinositol 3-kinase (PI3K) pathway, revealed increased PI3K activation in the airway field of smokers with lung cancer or bronchial premalignant lesions (PMLs) (13). These results suggest the airway field of injury reflects processes associated with a precancerous disease state; however, the molecular changes have not been adequately characterized.
This is an important shortcoming because bronchial PMLs are precursors of squamous cell lung carcinoma, yet effective tools to identify smokers with PMLs at highest risk of progression to invasive cancer are lacking. Several studies report loss of heterozygosity, chromosomal aneusomy, and aberrant methylation and protein expression in bronchial PMLs (14-23). These molecular events can give rise to histological changes that can be reproducibly graded by a pathologist prior to the development of invasive carcinoma. Autofluorescence bronchoscopy can be used to detect and sample PMLs, which have a prevalence of approximately 9% for moderate dysplasia and 0.8% for carcinoma in situ (CIS) (24-26). The presence of high grade PMLs (severe dysplasia or CIS) is a marker of increased lung cancer risk in both the central and peripheral airways indicating the presence of changes throughout the airway field (27, 28).
The molecular characterization of the airway field of injury in smokers with PMLs disclosed herein provides novel insights into the earliest stages of lung carcinogenesis and identifies relatively accessible biomarkers to guide early lung cancer detection and early intervention. Accordingly, disclosed herein are novel biomarkers and gene expression signatures and related assays and methods that are able to provide information about the precancerous disease state and if this pre-cancerous disease state is progressing and/or regressing. Such biomarkers and the related assays and methods are useful for monitoring the progression of premalignant or pre-cancerous conditions in a subject by obtaining (e.g., non-invasively obtaining) a biological sample of epithelial cells from the respiratory tract of the subject (e.g., bronchial or nasal epithelial cells). In certain aspects, alterations in gene expression observed in epithelial cells that are distal to the lung tissues (e.g., nasal or buccal epithelial cells) are concordant with changes in the bronchial epithelium.
The present inventions represent a significant advance in the detection and monitoring of individuals with premalignant lesions (PMLs), particularly in comparison to the standard of care auto-fluorescence bronchoscopy techniques which are less sensitive. In addition to detecting and monitoring of PMLs, the present inventions provide means of advancing the identification of chemoprevention agents, which historically has been bounded by the difficulty of discerning premalignant phenotypes through currently available means. The present inventions further provide means of using gene expression profiling as a surrogate end point that complements both histological and marker end points used today, such as Ki67.
The biomarkers and related methods and assays disclosed herein are based in part upon the finding of a strong correlation between PMLs and the alterations in gene expression in tissues that are physically distant from the site of disease (e.g., the nasal epithelium). It has further been found that these biomarkers strongly predict whether a suspected PML is pre-malignant. The biomarkers, assays and methods disclosed herein are characterized by the accuracy with which they can detect and monitor lung cancer and their non-invasive or minimally-invasive nature. In some aspects, the assays and methods disclosed herein are based on detecting differential expression of one or more genes in airway epithelial cells and such assays and methods are based on the discovery that such differential expression in airway epithelial cells are useful for identifying and monitoring PMLs in the distant lung tissue. Accordingly, the inventions disclosed herein provide a substantially less invasive method for diagnosis, prognosis and monitoring of lung cancer using gene expression analysis of biological samples comprising airway epithelial cells.
In contrast to conventional invasive methods, such as auto-fluorescence bronchoscopy, the assays and methods disclosed herein rely on expression of certain genes in a biological sample obtained from a subject. As the phrase is used herein, “biological sample” means any sample taken or derived from a subject comprising one or more airway epithelial cells. As used herein, the phrase “obtaining a biological sample” refers to any process for directly or indirectly acquiring a biological sample from a subject. For example, a biological sample may be obtained (e.g., at a point-of-care facility, a physician's office, a hospital) by procuring a tissue or fluid sample from a subject. Alternatively, a biological sample may be obtained by receiving the sample (e.g., at a laboratory facility) from one or more persons who procured the sample directly from the subject.
Such biological samples comprising airway epithelial cells may be obtained from a subject (e.g., a subject suspected of having one or more PMLs or that is otherwise at risk for developing lung cancer) using a brush or a swab. The biological sample comprising airway epithelial cells may be collected by any means known to one skilled in the art and, in certain embodiments, is obtained in a non-invasive or minimally-invasive manner. For example, in certain embodiments, a biological sample comprising airway epithelial cells (e.g., nasal epithelial cells) may be collected from a subject by nasal brushing. Similarly, nasal epithelial cells may be collected by brushing the inferior turbinate and/or the adjacent lateral nasal wall. For example, following local anesthesia with 2% lidocaine solution, a CYROBRUSH® (MedScand Medical, Malmoδ, Sweden) or a similar device, is inserted into the nare of the subject, for example the right nare, and under the inferior turbinate using a nasal speculum for visualization. The brush is turned (e.g., turned 1, 2, 3, 4, 5 times or more) to collect the nasal epithelial cells, which may then be subjected to analysis in accordance with the assays and methods disclosed herein.
In some embodiments, methods and assays disclosed herein may be assessed or performed on a biological sample that is obtained from a subject at a suspected site of a PML (e.g., premalignant bronchial lesion). In some embodiments, the suspected site is identified as having abnormal fluorescent during auto-fluorescence bronchoscopy, although the method of identifying the suspected site is not limited. In some embodiments, the methods and assays disclosed herein may be performed on a biopsy of a suspected PML as an alternative to, or in addition to, a histological examination of the biopsy.
In certain embodiments, the biological sample does not include or comprise bronchial airway epithelial cells. For example, in certain embodiments, the biological sample does not include epithelial cells from the mainstem bronchus. In certain aspects, the biological sample does not include cells or tissue collected from bronchoscopy. In some embodiments, the biological sample does not include cells or tissue isolated from a pulmonary lesion. In some embodiments, the biological sample does not include cells or tissue isolated from a PML.
To isolate nucleic acids from the biological sample, the airway epithelial cells can be placed immediately into a solution that prevents nucleic acids from degradation. For example, if the nasal epithelial cells are collected using the CYTOBRUSH, and one wishes to isolate RNA, the brush is placed immediately into an RNA stabilizer solution, such as RNALATER®, AMBION®, Inc. One can also isolate DNA. After brushing, the device can be placed in a buffer, such as phosphate buffered saline (PBS) for DNA isolation.
The nucleic acids (e.g., mRNA) are then subjected to gene expression analysis. Preferably, the nucleic acids are isolated and purified. However, if techniques such as microfluidic devices are used, cells may be placed into such device as whole cells without substantial purification. In one embodiment, airway epithelial cell gene expression is analyzed using gene/transcript groups and methods of using the expression profile of these gene/transcript groups in diagnosis and prognosis of lung diseases. In some embodiments, differential expression of the one or more genes determined with reference to the one or more of the 280 genes set forth in Table 3.
As used herein, the term “differential expression” refers to any qualitative or quantitative differences in the expression of the gene or differences in the expressed gene product (e.g., mRNA or microRNA) in the airway epithelial cells of the subject. A differentially expressed gene may qualitatively have its expression altered, including an activation or inactivation, in, for example, the presence of absence of cancer and, by comparing such expression in airway epithelial cell to the expression in a control sample in accordance with the methods and assays disclosed herein, and the presence or absence of PMLs may be determined and their progression or regression monitored.
In certain embodiments, the methods and assays disclosed herein are characterized as being much less invasive relative to, for example, bronchoscopy. The methods provided herein not only significantly increase the sensitivity or diagnostic accuracy of detecting and monitoring PMLs, but in certain aspects also make the analysis faster, much less invasive and thus much easier for the clinician to perform. In some embodiments, the likelihood that the subject has a PML or the likelihood that such a PML will progress to lung cancer is also determined based on the presence or absence of one or more secondary factors or diagnostic indicia of lung cancer, such as the subject's smoking history or status, or the results of previously performed imaging studies (e.g., chest CT scans). When the biomarkers, assays and methods of the present invention are combined with, for example, one or more relevant secondary factors (e.g., a subject's smoking history), the sensitivity and accuracy of detecting PMLs or their progression to lung cancer may be dramatically enhanced, enabling the detection of PMLs or their progression to lung cancer at an earlier stage, and by providing far fewer false negatives and/or false positives. As used herein, the phrase “secondary factors” refers broadly to any diagnostic indicia that would be relevant for determining a subject's risk of having or developing lung cancer. Exemplary secondary factors that may be used in combination with the methods or assays disclosed herein include, for example, imaging studies (e.g., chest X-ray, CT scan, etc.), the subject's smoking status or smoking history, the subject's family history and/or the subject's age. In certain aspects, when such secondary factors are combined with the methods and assays disclosed herein, the sensitivity, accuracy and/or predictive power of such methods and assays may be further enhanced. In some aspects, the methods and assays described herein are performed on a patient with a positive result in an imaging study (e.g., chest X-ray, CT scan, etc.). In some aspects, the methods or assays disclosed herein are used to confirm or rule out a positive result in an imaging study (e.g., chest X-ray, CT scan, etc.). In some aspects, the methods or assays disclosed herein are used to determine whether a positive result in an imaging study warrants a further invasive procedure (e.g., bronchoscopy), chemoprophylaxis, and/or chemotherapy.
The present inventors have discovered that PMLs and normal lung cells use different pathways to produce energy and survive and have harnessed this difference to develop the biomarker and related assays and methods disclosed herein. In some embodiments, the biological sample comprising the subject's airway epithelial cells (e.g., nasal or buccal epithelial cells) are analyzed for the expression of certain genes or gene transcripts corresponding to such metabolic pathways, either individually or in groups or subsets. In one embodiment, the inventions disclosed herein provide a group of genes corresponding to one or more pathways (e.g., metabolic pathways) that are significantly enriched in genes that are up- or down-regulated in the presence of PMLs (e.g., one or more pathways identified in Dataset 2) and that may be analyzed to determine the presence or absence of PMLs and/or their progression to lung cancer (e.g., adenocarcinoma, squamous cell carcinoma, small cell cancer and/or non-small cell cancer) from a biological sample comprising the subject's airway epithelial cells. For example, in certain aspects the biological sample may be analyzed to determine the differential expression of one or more genes from pathways involved in oxidative phosphorylation (OXPHOS), the electron transport chain (ETC), and mitochondrial protein transport to determine whether the subject has PMLs or is at risk of developing lung cancer. Other up-regulated pathways included DNA repair and the HIF1A pathway. Down-regulated pathways included the STATS pathway, the JAK/STAT pathway, IL-4 signaling, RAC1 regulatory pathway, NCAM1 interactions, collagen formation, and extracellular matrix organization.
In certain embodiments, the airway epithelial cells are analyzed using at least one and no more than 280 of the genes listed in Table 3. For example, about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 10-15, about 15-20, about 20-30, about 30-40, about 40-50, at least about 10, at least about 20, at least about 30, at least about 40, at least about 50, at least about 60, at least about 70, at least about 80, at least about 90, at least about 100, at least about 110, at least about 120, at least about 130, at least about 140, at least about 150, at least about 160, at least about 170, at least about 180, at least about 190, at least about 200, 210, 220, 230, 240, 250, 260, 270 or 275 or a maximum of the 280 genes as listed on Table 3.
Examples of the gene transcript groups useful in the diagnostic and prognostic assays and methods of the invention are set forth in Table 3. The present inventors have determined that taking any group that has at least about 5, 10, 15, 20, 25, 30, 40, 50, 60, 70, 80, 90, 100, 125, 150, 175, 200, 225, 250, 275 or more of the Table 3 genes provides a much greater PML detection sensitivity than chance alone. Preferably one would analyze the airway epithelial cells using more than about 20 of these genes, for example about 20-280 and any combination between, for example, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, and so on. In some instances, the present inventors have determined that one can enhance the sensitivity or diagnostic accuracy of the methods and assays disclosed herein by adding additional genes to any of these specific groups. For example, in certain aspects, the accuracy of such methods may approach about 70%, about 75%, about 80%, about 82.5%, about 85%, about 87.5%, about 88%, about 90%, about 92.5%, about 95%, about 97.5%, about 98%, about 99% or more by evaluating the differential expression of more genes from the set (e.g., the set of genes set forth in Table 3).
In some embodiments, the presence of PMLs or their progression/regression is made by comparing the expression of the genes or groups of genes set forth in, for example Table 3, by the subject's airway epithelial cells to a control subject or a control group (e.g., a positive control with confirmed PMLs or a confirmed diagnosis of lung cancer). In certain embodiments, an appropriate control is an expression level (or range of expression levels) of a particular gene that is indicative of the known presence of PMLs or a known lung cancer status. An appropriate reference can be determined experimentally by a practitioner of the methods disclosed herein or may be a pre-existing expression value or range of values. When an appropriate control is indicative of lung cancer, a lack of a detectable difference (e.g., lack of a statistically significant difference) between an expression level determined from a subject in need of characterization or diagnosis of lung cancer and the appropriate control may be indicative of lung cancer in the subject. When an appropriate control is indicative of the presence of PMLs or lung cancer, a difference between an expression level determined from a subject in need of characterization or determination of PMLs or diagnosis of lung cancer and the appropriate reference may be indicative of the subject being free of PMLs or lung cancer.
Alternatively, an appropriate control may be an expression level (or range of expression levels) of one or more genes that is indicative of a subject being free of PMLs or lung cancer. For example, an appropriate control may be representative of the expression level of a particular set of genes in a reference (control) biological sample obtained from a subject who is known to be free of PMLs or lung cancer. When an appropriate control is indicative of a subject being free of PMLs or lung cancer, a difference between an expression level determined from a subject in need of detection of PMLs or the diagnosis of lung cancer and the appropriate reference may be indicative of the presence of PMLs and/or lung cancer in the subject. Alternatively, when an appropriate reference is indicative of the subject being free of PMLs or lung cancer, a lack of a detectable difference (e.g., lack of a statistically significant difference) between an expression level determined from a subject in need of detection of PMLs or diagnosis of lung cancer and the appropriate reference level may be indicative of the subject being free of PMLs and/or lung cancer.
The control groups can be or comprise one or more subjects with a confirmed presence of PMLs, positive lung cancer diagnosis, a confirmed absence of PMLs or a negative lung cancer diagnosis. Preferably, the genes or their expression products in the airway epithelial cell sample of the subject are compared relative to a similar group, except that the members of the control groups may not have PMLs and/or lung cancer. For example, such a comparison may be performed in the airway epithelial cell sample from a smoker relative to a control group of smokers who do not have PMLs or lung cancer. The transcripts or expression products are then compared against the control to determine whether increased expression or decreased expression can be observed, which depends upon the particular gene or groups of genes being analyzed, as set forth, for example, in Table 3. In certain embodiments, at least 50% of the gene or groups of genes subjected to expression analysis must provide the described pattern. Greater reliability is obtained as the percent approaches 100%. Thus, in one embodiment, at least about 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 98%, 99% of the one or more genes subjected to expression analysis demonstrate an altered expression pattern that is indicative of the presence or absence of PMLs or lung cancer, as set forth in, for example, Table 3. Similarly, in one embodiment, at least about 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 98%, 99% of the one or more genes involved in a pathways set forth in Dataset 2 are subjected to expression analysis and demonstrate an altered expression pattern that is indicative of the subject's cancer status.
Any combination of the genes and/or transcripts of Table 3 can be used in connection with the assays and methods disclosed herein. In one embodiment, any combination of at least 5-10, 10-20, 20-30, 30-40, 40-50, 50-60, 60-70, 70-80, 80, 80-90, 90-100, 100-120, 120-140, 140-150, 150-160, 160-170, 170-180, 180-190, 190-200, 200-210, 210-220, 220-230, 230-240, 240-250, 250-260, 260-270 or 270-280 genes selected from the group consisting of genes or transcripts as shown in the Table 3.
The analysis of the gene expression of one or more genes may be performed using any gene expression methods known to one skilled in the art. Such methods include, but are not limited to expression analysis using nucleic acid chips (e.g. Affymetrix chips) and quantitative RT-PCR based methods using, for example real-time detection of the transcripts. Analysis of transcript levels according to the present invention can be made using total or messenger RNA or proteins encoded by the genes identified in the diagnostic gene groups of the present invention as a starting material. In certain aspects, analysis of transcript levels according to the present invention can be made using micronRNA. In the preferred embodiment the analysis is an immunohistochemical analysis with an antibody directed against proteins comprising at least about 10-20, 20-30, preferably at least 36, at least 36-50, 50, about 50-60, 60-70, 70-80, 80-90, 96, 100-180, 180-200, 200-250 or 250-280 of the proteins encoded by the genes and/or transcripts as shown in Table 3.
The methods of analyzing expression and/or determining an expression profile of the one or more genes include, for example, Northern-blot hybridization, ribonuclease protection assay, and reverse transcriptase polymerase chain reaction (RT-PCR) based methods. In certain aspects, the different RT-PCR based techniques are a suitable quantification method for diagnostic purposes of the present invention, because they are very sensitive and thus require only a small sample size which is desirable for a diagnostic test. A number of quantitative RT-PCR based methods have been described and are useful in measuring the amount of transcripts according to the present invention. These methods include RNA quantification using PCR and complementary DNA (cDNA) arrays (Shalon, et al., Genome Research 6(7):639-45, 1996; Bernard, et al., Nucleic Acids Research 24(8): 1435-42, 1996), real competitive PCR using a MALDI-TOF Mass spectrometry based approach (Ding, et al., PNAS, 100: 3059-64, 2003), solid-phase mini-sequencing technique, which is based upon a primer extension reaction (U.S. Pat. No. 6,013,431, Suomalainen, et al., Mol. Biotechnol. June; 15(2): 123-31, 2000), ion-pair high-performance liquid chromatography (Doris, et al., J. Chromatogr. A May 8; 806(1):47-60, 1998), and 5′ nuclease assay or real-time RT-PCR (Holland, et al., Proc Natl Acad Sci USA 88: 7276-7280, 1991).
The presently described gene expression profile can also be used to screen for subjects with confirmed PMLs to determine whether such subject are susceptible to or otherwise at risk for developing lung cancer. For example, a current smoker of advanced age (e.g., 70 years old) with PMLs may be at an increased risk for developing lung cancer and may represent an ideal candidate for the assays and methods disclosed herein. Moreover, the early detection of lung cancer in such a subject may improve the subject's overall survival. Accordingly, in certain aspects, the assays and methods disclosed herein are performed or otherwise comprise an analysis of the subject's secondary risk factors for developing cancer. For example, one or more secondary factors selected from the group consisting of advanced age (e.g., age greater than about 40 years, 50 years, 55 years, 60 years, 65 years, 70 years, 75 years, 80 years, 85 years, 90 years or more), smoking status, the presence of a lung nodule greater than 3 cm on CT scan and the time since the subject quit smoking. In certain embodiments, the assays and methods disclosed herein further comprise a step of considering the presence of any such secondary factors to inform the determination of whether the subject has PMLs or whether such PMLs are likely to progress to lung cancer.
As used herein, a “subject” means a human or animal Usually the animal is a vertebrate such as a primate, rodent, domestic animal or game animal. In certain embodiments, the subject is a mammal (e.g., a primate or a human). The subject may be an infant, a toddler, a child, a young adult, an adult or a geriatric. The subject may be a smoker, a former smoker or a non-smoker. The subject may have a personal or family history of cancer. The subject may have a cancer-free personal or family history. The subject may exhibit one or more symptoms of lung cancer or other lung disorder (e.g., emphysema, COPD). For example, the subject may have a new or persistent cough, worsening of an existing chronic cough, blood in the sputum, persistent bronchitis or repeated respiratory infections, chest pain, unexplained weight loss and/or fatigue, or breathing difficulties such as shortness of breath or wheezing. The subject may have a lesion, which may be observable by computer-aided tomography or chest X-ray. The subject may be an individual who has undergone a bronchoscopy or who has been identified as a candidate for bronchoscopy (e.g., because of the presence of a detectable lesion or suspicious imaging result). The terms, “patient” and “subject” are used interchangeably herein. In some embodiments, the subject is at risk for developing lung cancer. In some embodiments, the subject has PMLs or lung cancer and the assays and methods disclosed herein may be used to monitor the progression of the subject's disease or to monitor the efficacy of one or more treatment regimens.
In some embodiments, the methods and assays disclosed herein are useful for identifying subjects that are candidates for enrollment in a clinical trial to assess the efficacy of one or more chemotherapeutic agents. In certain aspects, the methods and assays disclosed herein are useful for determining a treatment course for a subject. For example, such methods and assays may involve determining the expression levels of one or more genes (e.g., one or more of the genes set forth in Table 3) in a biological sample obtained from the subject, and determining a treatment course for the subject based on the expression profile of such one or more genes. In some embodiments, the treatment course is determined based on a risk-score derived from the expression levels of the one or more genes analyzed. The subject may be identified as a candidate for a particular intervention or treatment based on an expression profile that indicates the subject's likelihood of having PMLs that will progress lung cancer. Similarly, the subject may be identified as a candidate for an invasive lung procedure (e.g., transthoracic needle aspiration, mediastinoscopy, lobectomy, or thoracotomy) based on an expression profile that indicates the subject has a relatively high likelihood of having PMLs or a high likelihood that such PMLs will progress to lung cancer (e.g., greater than 60%, greater than 70%, greater than 80%, greater than 90%). Conversely, the subject may be identified as not being a candidate for interventional therapy or an invasive lung procedure based on an expression profile that indicates the subject has a relatively low likelihood (e.g., less than 50%, less than 40%, less than 30%, less than 20%) of having PMLs or a low likelihood that such PMLs will progress to lung cancer. In some embodiments, a health care provider may elect to monitor the subject using the assays and methods disclosed herein and/or repeat the assays or methods at one or more later points in time, or undertake further diagnostics procedures to rule out PMLs or lung cancer. Also contemplated herein is the inclusion of one or more of the genes and/or transcripts presented in, for example, Table 3 into a composition or a system for detecting lung cancer in a subject. For example, any one or more genes and or gene transcripts from Table 3 may be added as a PML marker or lung cancer marker for a gene expression analysis. In some aspects, the present inventions relate to compositions that may be used to determine the expression profile of one or more genes from a subject's biological sample comprising airway epithelial cells. For example, compositions are provided that consist essentially of nucleic acid probes that specifically hybridize with one or more genes set forth in Table 3. These compositions may also include probes that specifically hybridize with one or more control genes and may further comprise appropriate buffers, salts or detection reagents. In certain embodiments, such probes may be fixed directly or indirectly to a solid support (e.g., a glass, plastic or silicon chip) or a bead (e.g., a magnetic bead).
The compositions described herein may be assembled into diagnostic or research kits to facilitate their use in one or more diagnostic or research applications. In some embodiments, such kits and diagnostic compositions are provided that comprise one or more probes capable of specifically hybridizing to up to 5, up to 10, up to 25, up to 50, up to 100, up to 200, up to 225, up to 250 or up to 280 genes set forth in Table 3 or their expression products (e.g., mRNA or microRNA). In some embodiments, each of the nucleic acid probes specifically hybridizes with one or more genes selected from those genes set forth in Table 3, or with a nucleic acid having a sequence complementary to such genes. A kit may include one or more containers housing one or more of the components provided in this disclosure and instructions for use. Specifically, such kits may include one or more compositions described herein, along with instructions describing the intended application and the proper use and/or disposition of these compositions. Kits may contain the components in appropriate concentrations or quantities for running various experiments.
The articles “a” and “an” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to include the plural referents. Claims or descriptions that include “or” between one or more members of a group are considered satisfied if one, more than one, or all of the group members are present in, employed in, or otherwise relevant to a given product or process unless indicated to the contrary or otherwise evident from the context. The invention includes embodiments in which exactly one member of the group is present in, employed in, or otherwise relevant to a given product or process. The invention also includes embodiments in which more than one, or the entire group members are present in, employed in, or otherwise relevant to a given product or process. Furthermore, it is to be understood that the invention encompasses all variations, combinations, and permutations in which one or more limitations, elements, clauses, descriptive terms, etc., from one or more of the listed claims is introduced into another claim dependent on the same base claim (or, as relevant, any other claim) unless otherwise indicated or unless it would be evident to one of ordinary skill in the art that a contradiction or inconsistency would arise. Where elements are presented as lists, (e.g., in Markush group or similar format) it is to be understood that each subgroup of the elements is also disclosed, and any element(s) can be removed from the group. It should be understood that, in general, where the invention, or aspects of the invention, is/are referred to as comprising particular elements, features, etc., certain embodiments of the invention or aspects of the invention consist, or consist essentially of, such elements, features, etc. For purposes of simplicity those embodiments have not in every case been specifically set forth in so many words herein. It should also be understood that any embodiment or aspect of the invention can be explicitly excluded from the claims, regardless of whether the specific exclusion is recited in the specification. The publications and other reference materials referenced herein to describe the background of the invention and to provide additional detail regarding its practice are hereby incorporated by reference.

EXAMPLES

Example 1

Patient Population

Bronchial airway brushings were obtained during autofluorescence bronchoscopy procedures between June 2000 and March 2011 from subjects in the British Columbia Lung Health Study at the British Columbia Cancer Agency (BCCA) (Vancouver, BC) (29) and between December 2009 and March 2013 from subjects in the High-Risk Lung Cancer-Screening Program at Roswell Park Cancer Institute (RPCI) (Buffalo, N.Y.) (detailed cohort information in the Methods section below). Premalignant Lesions were sampled (if present) using endobronchial biopsy, graded by a team of pathologists at BCCA or RPCI, and the worst histology observed was recorded. Bronchial brushes of normal-appearing epithelium from 84 BCCA subjects (1 brush per subject) with and without PMLs were selected to undergo mRNA-Seq while ensuring balanced clinical covariates. Fifty-one bronchial brushes of normal-appearing epithelium from 23 RPCI subjects were also profiled by mRNA-Seq (18 subjects had 2 procedures, and 5 subjects had 3 procedures). The RPCI samples were utilized in biomarker validation to calculate changes in the biomarker score between sequential procedures. Sets of samples were classified as stable/progressive if the worst histological grade at the second time point for a given patient remained the same or worsened, and regressive if the worst histological grade at the second time point improved. The Institutional Review Boards (IRBs) of all participating institutions approved the study and all subjects provided written informed consent.

RNA-Seq Library Preparation, Sequencing and Data Processing

Total RNA was extracted from bronchial brushings using miRNeasy Mini Kit (Qiagen). Sequencing libraries were prepared from total RNA samples using Illumina® TruSeq® RNA Kit v2 and multiplexed in groups of four using Illumina® TruSeq® Paired-End Cluster Kit. Each sample was sequenced on the Illumina® HiSeq® 2500 to generate paired-end 100 nucleotide reads. Demultiplexing and creation of FASTQ files were performed using Illumina CASAVA v1.8.2. For the BCCA samples, reads were aligned to hg19 using TopHat v2.0.4. The insert size mean and standard deviation were determined using the alignments and MISO (32). Reads were realigned using TopHat and the insert size parameters. Alignment and quality metrics were calculated using RSeQC v2.3.3. Gene count estimates were derived using HTSeq-count v0.5.4 (33) and the Ensembl v64 GTF file. Gene filtering was conducted on normalized counts per million (cpm) calculated using R v3.0.0 and edgeR v3.4.2 using a modified version of the mixture model in the SCAN. UPC Bioconductor package (34). A gene was included in downstream analyses if the mixture model classified it as “on” (i.e. “signal”) in at least 15% of the samples. For the RPCI samples, gene counts were computed using RSEM (v1.2.1) (30) and Bowtie (v1.0.0) (31) with Ensembl 74 annotation. The data is available from NCBI's Gene Expression Omnibus (GEO) using the accession ID GSE79315.

Data Analysis for the BCCA Samples

Sample and gene filtering yielded 13,870 out of 51,979 genes and 82 samples (n=2 excluded due to quality or sex annotation mismatches) for analysis. Data from Beane et al. (3) was used to predict the smoking status of the 82 samples (Dataset 1, FIG. 6 and Methods) used in all further analysis. Airway brushings were dichotomized into two groups: samples with no evidence of PMLs (samples with no abnormal fluorescing areas or biopsies having normal or hyperplasia histology, n=25); and samples with evidence of PMLs (biopsies having mild, moderate, or severe dysplasia, n=50). Brushes with a worst histology of metaplasia (n=7) were excluded from the dichotomized groups. The limma (35), edgeR (36) and sva packages (37) were used to identify differentially expressed genes associated with presence of PMLs using normalized voom-tranformed (38) data and surrogate variable analysis using the first 7 surrogate variables (Table 51). Gene set enrichment analyses were conducted using ROAST (39) and GSEA (40), and GSVA (41). The Molecular Signatures Database (MSigDb) v4 Entrez ID Gene Sets were converted to Ensembl IDs using BioMart. Additional gene sets were created from CEL files or RNA-Seq counts from The Cancer Cell Line Compendium (CCLE), SCC tumor and adjacent normal tissue from TCGA, GSE19188, GSE18842, and GSE4115 (Supplemental Methods).

Cell Culture

The human bronchial epithelial biopsy cell cultures (Table S2) were obtained from the Colorado Lung SPORE Tissue Bank and cultured in Bronchial Epithelial Growth Media (BEGM). Human non-small cell lung cancer (NSCLC) cell lines were purchased from ATCC and short tandem repeat (STR) profiles were verified at the time of use by the Promega Gene Print® 10 system at the Dana Faber Cancer Institute. H1299, H2085 and SW900 cells were cultured in RPMI supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, and H2085 cells were cultured in ALC-4 media. All cells were grown in a 37° C. humidified incubator with 5% CO₂.

Bioenergetics Studies

Oxygen consumption rates (OCR) and extracellular acidification rates (ECAR) were measured using the XF96 Extracellular Flux Analyzer instrument (Seahorse Bioscience Inc). Briefly, approximately 30,000 cancer cells/well or approximately 40,000 bronchial epithelial biopsy cells/well (higher numbers due to slow growth rate) were seeded on XF96 cell culture plates and grown overnight. Prior to running the assay, media was replaced with Seahorse base media (2 mM (milimole/L) L-glutamine) and placed at 37° C. and 0% CO₂for approximately 30 minutes. The XF Cell Mito Stress Test kit and protocol were utilized to examine mitochondrial function. Measurements were taken every 5 minutes over 80 minutes. To modulate mitochondrial respiration, 504 oligomycin, 1 μM FCCP and 504 antimycin A were used. Prism software v6 was used to calculate t-statistics for baseline OCR comparisons and a 2-way ANOVA was conducted to compare OCR and ECAR measurements.

Mitochondrial Enumeration Using Flow Cytometry

Using an established protocol (40), cell cultures (5×10⁵cells/10 cc dish of bronchial biopsy cultures and cancer cell cultures) were grown overnight and exposed to 120 uM MitoTracker Green FM in media free of FBS for 30 min at 37° C. humidified incubator with 5% CO₂. Cells were subsequently collected, washed in PBS and resuspended in 0.5 mL PBS-EDTA and 1 uL of propidium iodide (PI) was added to distinguish live/dead cells. MitoTracker FM and PI were measured using a BD LSRII flow cytometer and BD FACS Diva software (6.2.1). Data was analyzed using FlowJo (10.2), gating out doublets and dead cells, and normalizing mean fluorescence to the number of cell counts.

Immunohistochemistry

Formalin-fixed, paraffin-embedded (FFPE) sections of human PMLs sampled from high-risk subjects undergoing screening for lung cancer were provided by RPCI as part of an IRB-approved study detailed below (Table S3). Dr. Candace Johnson at RPCI provided the PIPE lung sections from the N-nitroso-tris-chloroethylurea (NTCU) mouse model of lung SCC, from mice treated with 25 ml of 40 mmol/L NTCU for 25 weeks in accordance with the Institutional Animal Care and Use Committee approved protocol (42). Antibody dilutions and immunohistochemistry methods were detailed in the Supplemental Methods. Briefly, slides were de-paraffinized and rehydrated. For antigen retrieval, slides were heated in citrate buffer. Slides were subsequently incubated in primary antibody (Translocase of the Outer Mitochondrial Membrane 22 (TOMM22): mouse tissue 1:300 and human 1:1,200 (Abcam), and Cytochrome C Oxidase subunit IV (COX4I1): mouse tissue 1:500 and human 1:5,000 (Abcam)) diluted in 1% Bovine Serum Albumin (BSA). Signal was amplified using an ABC kit (Vector Labs). To reveal endogenous peroxidase activity, slides were incubated in a 3,3′-Diaminobenzidine (DAB) solution. Slides were rinsed, counterstained with hematoxylin, dehydrated in graded alcohol followed by xylene and cover slipped.

Biomarker Development and Validation

A gene expression biomarker discovery pipeline was developed to test thousands of parameter combinations (6,160 predictive models) to identify a biomarker capable of distinguishing between samples from subjects with and without PMLs. Samples were first assigned by batch (sequencing lane) to either a discovery set (n=58) or a validation set (n=17), and the validation set was excluded from biomarker development (FIG. S2 and Supplemental Methods). The biomarker was developed using subsets of the discovery set established by randomly splitting the samples into training (80%, n=46) and test (20%, n=12) sets 500 times. Model performance was assessed using standard metrics for both the training and test sets (Supplemental Methods). The biomarker pipeline was also used to develop biomarkers for sex and smoking status as well as randomized class labels for all phenotypes (serving as positive and negative controls, respectively). A final model (biomarker) was selected (Supplemental Methods) and its ability to distinguish between samples with and without PMLs was tested in a validation set (n=17). In addition, using the bronchial brushings collected longitudinally from subjects at RPCI, we tested whether or not differences in biomarker scores over time were reflective of progression of PMLs (n=28 matched time point pairs) (Supplemental Methods).

Example 2

Results

Subject Population

The study design used 126 bronchial brushings obtained via autofluorescence bronchoscopy at the BCCA and RPCI for differential gene expression and pathway analysis, as well as for biomarker development and validation (FIG. 1). A dataset consisting of samples collected from BCCA subjects with (n=50) and without (n=25) PMLs (n=25) was used to derive a gene expression signature associated with the presence of dysplastic PMLs. Important clinical covariates such as COPD and reported smoking history as well as alignment statistics from the mRNA-Seq data were not significantly different between the two groups (Table 1 and Table 2). For biomarker development, the 75 BCCA samples were split by batch and used in biomarker discovery (n=58) and validation (n=17) (Tables S4 and S5). The change in biomarker score as a predictor of progression of PMLs was then tested in the 51 RPCI samples (Tables S5 and S6).
Transcriptomic Alterations in the Airway Field of Injury Associated with the Presence of PMLs
The present inventors identified 280 genes significantly differentially expressed between subjects with and without PMLs (FDR<0.002, FIG. 2). Utilizing the Molecular Signatures Database v4 (MSigDB) canonical pathways, the present inventors identified 170 pathways significantly enriched in genes up- or down-regulated in the presence of PMLs using ROAST (39) (FDR<0.05, Dataset 2). Pathways involved in oxidative phosphorylation (OXPHOS), the electron transport chain (ETC), and mitochondrial protein transport were strongly enriched among genes up-regulated in the airways of subjects with PMLs. Other up-regulated pathways included DNA repair and the HIF1A pathway. Down-regulated pathways included the STATS pathway, the JAK/STAT pathway, IL4 signaling, RAC1 regulatory pathway, NCAM1 interactions, collagen formation, and extracellular matrix organization.

OXPHOS is Increased in PML Cell Cultures and Biopsies of Increasing Severity

The ETC and OXPHOS pathways, which involve genes distributed between the complexes I-IV of the ETC and ATP synthase, were highly activated in the airway field in the presence of PMLs. The present inventors wanted to determine if the functional activity of these pathways was similarly altered in PMLs compared to normal tissue. Cellular bioenergetics were conducted by measuring oxygen consumption rate (OCR) as a measure of ETC/OXPHOS and extracellular acidification rate (ECAR) as a measure of glycolysis (anerobic respiration) and MitoTraker Green FM as a measure of mitochondrial content in primary cell cultures derived from bronchial biopsies. Additionally, the present inventors performed immunohistochemistry of select OXPHOS-related genes in mouse and human dysplastic lesions and normal tissue to measure protein levels.
The present inventors established a significant concordance between ETC/OXPHOS gene expression and cellular bioenergetics in NSCLC cell lines (FIGS. 7A-7F). Next, using primary cell cultures derived from normal to severe dysplastic tissue (Table S2), the present inventors observed that the mean baseline OCR values were 2.5 fold higher in the cultures from PMLs compared to controls (p<0.001, FIG. 3A). Baseline ECAR values were also higher in PML cultures compared to controls, but to a lesser extent (1.5 fold, p<0.001), reflecting predictions based on mRNA-Seq field data (FIGS. 7G-7H). There was a greater reduction in OCR in PMLs immediately following oligomycin treatment (p<0.001) suggesting an increased dependence on OXPHOS for ATP production to meet energetic demands. In addition, the mean spare respiratory capacity following the release of the proton gradient was elevated by approximately 1.5 fold in the PML cultures compared to controls indicating increased ability to respond to energy demands (43). Lastly, treatment with antimycin A resulted in a greater reduction of OCR in PML cultures (p<0.001, FIG. 3B), suggesting that oxygen consumption in the lesions is dependent on increased ETC components in complex III. No significant changes to ECAR were detected in response to mitochondrial perturbations. Furthermore to examine if the increased OXPHOS was a result of increased mitochondrial biogenesis in PML cultures, cells were incubated with MitoTraker FM to stain for mitochondria content and fluorescence enumerated using flow cytometry revealed no significant difference between PML and controls (p=0.15, FIG. 3C-D).
Additionally, the present inventors found elevated protein levels of Translocase of the Outer Mitochondrial Membrane 22 (TOMM22) and Cytochrome C Oxidase subunit IV (COX4I1) in low/moderate grade dysplastic lesions compared to normal tissue (FIG. 3C) using tissues from human bronchial biopsy FFPE sections (Table S3) and whole lung sections from the NTCU mouse model of SCC. The results suggest that PMLs are more ETC- and OXPHOS-dependent and express OXPHOS-related proteins at higher levels compared to normal tissue.

PML-Associated Gene Expression Alterations in the Airway Field are Involved in Lung Squamous Cell Carcinogenesis

To further extend the connection between the airway field and PMLs, the present inventors examined the relationship between PML-associated genes in the airway field and other lung cancer-related datasets. The present inventors identified genes differentially expressed between lung tumor tissue (primarily squamous) and normal lung tissue in three different datasets (TCGA, GSE19188, and GSE18842). Genes associated with lung cancer in all datasets were significantly (FDR<0.05) enriched by GSEA, concordantly with gene expression changes associated with the presence of PMLs in the field (FIG. 4A and Dataset 3). Extending beyond the lung tumor, similar enrichment (FDR<0.05) was found using early, stepwise, and late gene expression changes in SCC identified by Ooi et al. (44) (FIG. 4B and Dataset 3) and among genes associated with lung cancer in the airway field of injury (GSE4115, FIG. 4C and Dataset 3). These results support the concept that early events in lung carcinogenesis can be observed throughout the respiratory tract, even in cells that appear cytologically normal.

Development and Validation of a Biomarker for PML Detection and Monitoring

The airway brushings from BCCA subjects with and without PMLs were leveraged to build a biomarker predictive of the presence of PMLs. The biomarker consisted of 200 genes (of which 91 overlapped with the gene signature in FIG. 2) and achieved a ROC-curve AUC of 0.92, sensitivity of 0.75 (9/12 samples with PMLs predicted correctly), and specificity of 1.00 (5/5 samples without PMLs predicted correctly) in independent validation samples (n=17, FIG. 5A). In addition, the biomarker was used to score an independent set of longitudinally collected bronchial brushings from RPCI subjects (FIG. 1). Biomarker scores were calculated for each sample, and the difference in biomarker scores between sequential procedures (n=28 time point pairs, Supplemental Methods) was predictive of whether the worst PML histology observed during the baseline procedure regressed or whether it was stable or progressed with an AUC of 0.75 (FIG. 5B).

Biomarker Predicts Dysplasia Status in Bronchial Biopsies

Abnormal fluorescing areas were biopsied during auto-fluorescence bronchoscopy of 91 subjects. Biopsies from 47 of the subjects were determined to be premalignant legions (severe, moderate or mild dysplasia) via histology. Biopsies from 44 of the subjects were determined to be normal (normal or hyperplasia) via histology. The ability of the biomarker to predict dysplasia status was assessed. FIG. 9 shows an ROC curve demonstrating the performance of the biomarker in distinguishing between premalignant lesion biopsies (severe=8, moderate=25, and mild dysplasia=14) and biopsies with normal histology (normal=24 and hyperplasia=20). Biomarker achieved AUC of 72% (with a 62%-83% confidence interval), sensitivity of 81% (38 of 47 dysplastic biopsies predicted correctly), and specificity of 66% (29 of 44 normal biopsies predicted correctly).

Discussion

In the foregoing studies, the present inventors identified a PML-associated gene expression signature in cytologically normal bronchial brushings and characterized the biological pathways that are dysregulated in the airway field of injury. The present inventors established that the PML-associated airway field harbors alterations observed in PMLs and in SCC. This evidence motivated the development of a biomarker that reflects the presence of PMLs and their outcome over time. The findings presented herein provide novel insights into the earliest molecular events associated with lung carcinogenesis and have the potential to impact lung cancer prevention by providing novel targets (e.g., OXPHOS) and potential biomarkers for risk stratification and monitoring the efficacy of chemoprevention agents.
The first major finding of the foregoing studies was the identification of a PML-associated field of injury. The most significantly enriched pathways among up-regulated genes in subjects with PMLs were OXPHOS, ETC, and mitochondrial protein transport. These pathways efficiently generate energy in the form of ATP by utilizing the ETC in the mitochondria. During cancer development, energy metabolism alterations are described as an increase in glycolysis and suppression of OXPHOS, known as the Warburg effect (45); however, recent studies demonstrate that OXPHOS is maintained in many tumors and can be important for progression (46). The present inventors wanted to assay for OXPHOS activation in PMLs as it may support PML progression by generating reactive oxygen species (ROS) that can induce oxidative stress, increase DNA damage, and HIF-1α pathway activation (pathways observed in our analysis).
The present inventors observed increases in both the basal OCR and the spare respiratory capacity in the PML biopsies, suggesting that PML-derived cell cultures are more ETC and OXPHOS dependent that the non-PML cultures. The present inventors also demonstrated increases in the presence of mitochondria and ETC activity marked by positive TOMM22 and COX IV staining associated with increasing PML histological grade. Several members of the mitochondrial protein import machinery (46) were significantly up-regulated (FDR<0.05) in airways with PMLs including members of the TOM complex (TOMM22, TOMM7, and TOMM20) and TIM23 complex (TIMM23, TIMM21, and TIMM17A). We observed positive staining of TOMM22 with increasing PML grade, suggesting that increased import of precursor proteins from the endoplasmic reticulum may be required to meet the energy demands of PMLs. Measurements of mitochondrial content indicated no significant differences between the normal and PML-derived cultures, and transcriptional levels of PPARGC1A, associated with mitochondrial biogenesis, were not different between subjects with and without PML indicating that increases in OXPHOS are likely independent of mitochondrial number (47-49). Increases in OXPHOS have been demonstrated to be associated with PML progression in Barret's esophagus and esophageal dysplasia (47), cervical dysplasia (48), and the dysplastic lesions that precede oral SCC (49). Collectively, these data suggest that the OXPHOS pathway may be a target for early intervention. Pre-clinical studies in the NTCU mouse model of lung SCC demonstrate the potential for targeting mitochondrial respiration by using the natural product honokiol to inhibit tumor development (50). Further investigations into the role of cellular energy metabolism in the development and progression of PMLs are needed to fully understand how to best target it for intervention in lung cancer.
Additionally, the present inventors extended the connection between the PML-associated airway field and PMLs beyond the OXPHOS pathway to processes associated with squamous cell lung carcinogenesis. By examining gene sets from multiple external studies representative of lung cancer-related processes occurring in the tumor, adjacent to the tumor, and in the upper airway, significant concordant relationships were found between the PML-associated field and processes associated with SCC tumors. Genes are similarly altered in these varied cancer-associated contexts and thus tissues in the field both adjacent to and far away from the tumor may reflect basic processes and mechanisms of lung carcinogenesis such as DNA damage as hypothesized earlier.
These observations motivated the present inventors to pursue the most translational aspect of this study, a biomarker that can detect PMLs and monitor their progression over time. The 200-gene biomarker, measured in the cytologically normal bronchial airway, achieved high performance detecting the presence of PMLs in a small test set (AUC=0.92). This biomarker may increase the sensitivity of bronchoscopy in detecting the presence of PMLs (which can be difficult to observe under white light), and thus improve identification of high-risk smokers that should be targeted for aggressive lung cancer screening programs. Additionally, the biomarker may offer wider clinical utility in early intervention trials by serving as an intermediate endpoint of efficacy (beyond Ki-67 staining for proliferation, and changes in biopsy histology). Towards this goal, the present inventors demonstrated that the change in biomarker scores over time reflected contemporaneous regressive or progressive/stable disease (AUC=0.75). This result suggests that the airway field of injury in the presence of PMLs is dynamic and that capturing the gene expression longitudinally may allow for further stratification of high-risk subjects. The potential clinical utility of the biomarker is further supported by recent work demonstrating a significant association between the development of incident lung squamous cell carcinoma and the frequency of sites that persist or progress to high-grade dysplasia (24).
Further development and testing in a larger cohort is needed to confirm the biomarker's performance, utility, and ability to predict future PML progression or regression. Additionally, longitudinal and spatial sampling would provide a greater understanding of the dynamic relationship between the normal epithelium and the PMLs as they regress or progress to SCC. Longitudinal studies would allow for more accurate characterization of the time intervals needed to observe gene expression dynamics both in the PMLs and in the airway field of injury. Spatial sampling throughout the respiratory tract, including the more accessible nasal airway that shares the tobacco-related injury with the bronchial airways (51), would allow for evaluation of the impact of distance between the PMLs and the brushing site, the range of PML histologies, and the multiplicity of PMLs that can be present simultaneously in a patient and influence the PML-associated airway field.
Despite these challenges and opportunities for future work, the present inventors have comprehensively profiled gene expression changes in airway epithelial cells in the presence of PMLs that suggest great clinical utility. Moving therapeutics and detection strategies towards an earlier stage in the disease process via molecular characterization of premalignant disease holds great promise (52, 53), and this study represents an important step towards a precision medicine approach to lung cancer prevention.
Materials and Methods
Software versions referenced

Data Processing

Illumina CASAVA v1.8.2

TopHat v2.0.4

RSeQC v2.3.3

HTSeq-count v0.5.4

R v3.0.0

edgeR v3.4.2

RSEM v1.2.1

Bowtie v1.0.0

Data Analysis

Limma v3.18.13

edgeR v3.4.2
sva v3.6.0

GSVA v1.10.3

Gene Expression-Based Prediction of Smoking Status

Microarray data from Beane et al. (3) Gene Expression Omnibus [GEO] (54) Accession Number GSE7895) was re-analyzed using using Robust Multi-array Average (RMA) (54) and the Ensembl CDF file v16.0.0 file website (brainarray.mbni.med.umich.edu/Brainarray/Database/CustomCDF/16.0.0/ensg.asp). The R package (35) was used to identify genes differentially expressed between current (n=52) and never (n=21) smokers, using the linear model presented in the paper additionally correcting for quality covariates (NUSE and RLE). Ninety-four genes (FDR<0.001) were differentially expressed between current and never smokers. The weighted voting algorithm (55) was trained on z-score normalized microarray data (n=73) across the 94 genes and used to predict smoking status in z-scored log 2-transformed counts per million (cpm) from the 82 mRNA-Seq samples.

Processing of Publically Available Datasets

Cancer Cell Line Compendium (CCLE). The Entrez ID gene expression file labeled 10/18/2012 and the sample information file were downloaded from CCLE website (broadinstitute.org/ccle/home). After matching the sample annotation to the expression file, we used ComBat (56) to adjust the data for batch effects (n=14 batches across 1019 samples). After batch correction, the lung cell lines (n=186) were selected and GSVA was used to calculate a pathway enrichment score for each lung cell line for the following pathways: KEGG oxidative phosphorylation, KEGG glycolysis gluconeogenesis, BioCarta glycolysis, and Reactome glycolysis. The GSVA scores for the glycolysis pathways were averaged per sample.
The Cancer Genome Atlas (TCGA). RSEM gene-level (Entrez IDs) counts derived from RNA-Seq data were downloaded from the TCGA data portal on Aug. 27, 2013, for lung squamous cell carcinomas and adjacent matched control tissue (n=100 samples from n=50 subjects). After applying the mixture model referenced in the paper, 14,178 out of 20,531 genes were expressed as signal in at least 15% of samples (n=15). Differential gene expression between tumor and adjacent normal tissue was determined using limma and voom-transformed data (38) via a linear model with cancer status as the main effect and a random patient effect modeled using the duplicateCorrelation function. Gene sets containing the top 200 up- and down-regulated differentially expressed genes associated with cancer status were used as input for GSEA.
Microarray Data. CEL files for GSE19188 and GSE18842 were downloaded from GEO and processed using Robust Multi-array Average (RMA) (54) and the Ensembl Gene CDF v16.0.0 file website (brainarray.mbni.med.umich.edu/Brainarray/Database/CustomCDF/16.0.0/ensg.asp). Samples with a median RLE greater than 0.1 or a median NUSE greater than 1.05 were excluded, yielding n=146 samples for GSE19188 and n=82 samples for GSE18842. For GSE19188, differential gene expression between squamous cell tumors (n=23) and normal lung tissue (n=64) was conducted using limma and a linear model that included RLE and NUSE covariates. For GSE18842, paired normal and tumor tissue from the same subjects (n=37 subjects, n=74 samples) were selected, and differential gene expression was conducted in an analogous manner as described above for TCGA, additionally correcting for RLE and NUSE metrics.
CEL files for GSE4115 were processed using RMA and the CDF file above. The n=164 samples described in Spira et al. (9), were used to determine genes differentially expressed in airway brushings from subjects with and without lung cancer, using limma and a linear model with terms for cancer status, RLE, NUSE, smoking status, and pack-years. Gene sets containing the top 200 up- and down-regulated differentially expressed genes associated with cancer status were used as input for GSEA.

Immunohistochemistry

Slides were de-paraffinized, rehydrated, and heated in citrate buffer for antigen retrieval. Slides were treated with 3% H₂O₂(in methanol) to block endogenous peroxidases, incubated in 10% normal goat serum, and primary antibody (TOMM22: mouse tissue 1:300 and human 1:1,200 (Abcam), and COX IV: mouse tissue 1:500 and human 1:5,000 (Abcam)) diluted in 1% BSA. Signal was amplified using an ABC kit (Vector Labs). Slides were next incubated in a 3,3′-Diaminobenzidine (DAB) solution to reveal endogenous peroxidase activity, rinsed, counterstained with hematoxylin, dehydrated in graded alcohol followed by xylene, and cover slipped.

Biomarker Development

Upstream gene filtering. In order to provide cross-platform compatibility, the present inventors ran the biomarker discovery and validation pipelines using 11,926 genes commonly present on the RNA-Seq platform (Illumina HiSeq 2500 used with Ensembl v64 GTF) and two microarray platforms (Affymetrix GeneChip Human Gene 1.0 ST Array used with custom ENSG Homo sapiens CDF from Brainarray v11 and Affymetrix Human Genome U133A Array used with custom ENSG Homo sapiens CDF from Brainarray v16).
Data generation and summarization. Samples (n=75) were run across 4 flow cells (4 batches), and samples run in batches 1, 2, and 3 (n=58) were assigned to a discovery set, while the remaining samples (n=17) were used as an independent validation set and not included in the biomarker development. Alignments and gene level summarization were conducted as described in the paper methods. Alignment and quality metrics were calculated using RSeQC (v2.3.3) (57). Using the gene body measure computed by RSeQC, a ratio between the average read coverage at 80% of the gene length and the average coverage at 20% of the gene length was derived as an additional quality metric (gb-ratio) to assess 3′ bias per sample. The metric was highly correlated with a surrogate variable applied in the identification of differentially expressed genes, and was used as a quality control metric in the biomarker pipeline.
Biomarker discovery pipeline. The biomarker discovery pipeline has been outlined generally above. A graphical representation of data flow as well as processing and analysis steps is provided in FIG. 8. Each computational step outlined is detailed in the following sections.
Balancing signature. The present inventors tested gene signatures consisting either of an equal or unequal number of genes up- and down-regulated in subjects with dysplastic lesions.
Input data preprocessing. The present inventors tested 3 input data types. HTSeq-count (v0.5.4) (33) was used to derive gene count estimates (raw counts). In addition, Cufflinks (v2.0.2) (58) was used to derive reads per kilobase per million mapped reads (RPKM) using BAM files containing only properly paired reads. The present inventors also calculated log 2-transformed counts per million (CPM) by applying edgeR (v3.8.6) (36) to raw counts using the “TMM” method (weighted trimmed mean of M-values (59)).
Gene filtering. Signal-based gene filtering was conducted as described in detail above (Methods). In short, a gene was included in downstream analyses if the mixture model classified it as “on” in at least 1%, 5%, 10% or 15% of the samples. For CPM input data type, the present inventors recalculated CPM values using raw counts after filtering out genes.
Feature selection. To identify genes differentially expressed (DE) between samples with and without premalignant lesions (PMLs), the present inventors applied several algorithms to our filtered dataset. The algorithms used were as follows:
(1) edgeR: The present inventors applied the edgeR package (v3.8.6) (46) to raw counts only. After calculating normalization factors (calcNormFactors) and estimating common (estimateGLMCommonDisp) and tagwise (estimateGLMTagwiseDisp) dispersion factors, we identified DE genes associated with the presence of PMLs using a generalized linear model, correcting for sex, COPD status, and smoking status covariates. For balanced signatures, the sign of the log 2-fold change of expression between conditions determined gene directionality. For all models regardless of balancing, gene importance was defined by FDR-adjusted p-value from likelihood ratio tests (glmLRT).
(2) edgeRgb: The present inventors used the edgeR package as described in #1, additionally correcting for gb-ratio (described above in the Data generation and summarization section).
(3) lm: The present inventors applied the limma package (v3.22.7) (35) to CPMs, RPKMs, or voom-transformed raw counts (38). Voom transformation was applied using a linear model, adjusting for sex, COPD status, and smoking status covariates, after calculating normalization factors. The same model was used to identify DE genes associated with the presence of PMLs. For balanced signatures, the sign of the moderated t-statistic obtained via eBayes and topTable determined gene directionality. For all models regardless of balancing, gene importance was defined by the magnitude of the t-statistic.
(4) lmgb: The present inventors used the limma package as described in #3, additionally correcting for gb-ratio (described above in the Data generation and summarization section).
(5) glmnet: The inventors applied the glmnet package (v1.9-8) (60) to CPMs, RPKMs, or voom-transformed raw counts (as in #3) to identify DE genes associated with the presence of PMLs. For balanced signatures, gene directionality was determined by the sign of the t-statistic obtained via limma by running a linear model described in #3. The inventors carried out the following series of steps using all genes for unbalanced signatures and separately using up- and down-regulated genes for balanced signatures: First, RPKMs and CPMs were z-score normalized, while raw counts were voom-transformed. Then, due to the binary character of our response variable (dysplasia status), a logistic regression model was fit using the binomial distribution family and elastic net mixing parameter α=0.5 (indicating a tradeoff between ridge and lasso regressions). The standardize option was set to FALSE, causing the coefficients to be returned on the original scale, thus allowing their magnitude to be interpreted as gene importance. Next, a range of regularization parameters λ was generated via leave-one-out cross-validation (nfolds=46), and the λ giving the minimum mean cross-validated error (lambda.min) was chosen to estimate the coefficients. Finally, DE genes were defined as having non-zero coefficients and then sorted by importance based on the coefficients' magnitude.
(6) randomForest: The inventors applied the randomForest package (v4.6-12) (61) to CPMs, RPKMs, and voom-transformed raw counts (as in #3), setting the number of trees (ntree) to 100 and importance to TRUE. For balanced signatures, the sign of the t-statistic as described in #5 determined gene directionality. For all models regardless of balancing, gene importance was determined by the magnitude of the importance variable, defined as the mean decrease in accuracy over both conditions.
(7) DESeq: The inventors applied the DESeq package (v1.18.0) (62) to unmodified raw counts only. DE analysis to find genes associated with the presence of PMLs included data normalization (estimation of the effective library size), variance estimation, and inference for two experimental conditions, as outlined in the DESeq package vignette (bioconductor.org/packages/3.3/bioc/vignettes/DESeq/inst/doc/DESeq.pdf). For balanced signatures, the sign of the log 2-fold change of expression between the two conditions determined gene directionality. For all models regardless of balancing, gene importance was defined by FDR.
(8) SVA: The inventors applied the sva package (v3.12.0) (37) to CPMs, RPKMs, or voom-transformed raw counts. Raw counts were voom-transformed using a linear model including only dysplasia status as the predictor variable. The number of surrogate variables (SVs) not associated with dysplasia status was estimated using the default approach of Buja and Eyuboglu (63) (“be” method). SVs were then identified using the empirical estimation of control probes (“irw” method), and up to 5 were added as covariates in the linear model (limma package). The adjusted model was then used to once again voom-transform raw counts, and subsequently fitted to identify DE genes associated with the presence of PMLs. For balanced signatures, the sign of the moderated t-statistic obtained via topTable determined gene directionality. For all models regardless of balancing, gene importance was defined by the magnitude of the t-statistic.
(9) pAUC (partial AUC) (64): The present inventors applied the rowpAUCs function in the genefilter package (v1.48.1) (65) to CPMs, RPKMs, or voom-transformed raw counts (as in #3). The inventors used 10 class label permutations and a sensitivity cutoff of 0.1 for a specificity range of 0.9-1. For balanced signatures, the sign of the moderated t-statistic obtained via limma's topTable determined gene directionality. For all models regardless of balancing, gene importance was defined by the magnitude of the t-statistic.
Gene signature size. After the feature selection step, the inventors selected the top scoring 10, 20, 40, 60, 80, 100, or 200 genes, making sure that for balanced signatures, half originated from an ordered list of up-regulated genes, and half from an ordered list of down-regulated genes.
Prediction method. For each set of genes, multiple prediction methods were applied to predict dysplasia status (presence of PMLs) in a training set of 46 samples and a test set of 12 samples. These training and test set samples differed in each iteration, which resulted from randomly splitting the 58 discovery set samples (FIG. 8). The following prediction methods were used:
1. glmnet: The inventors used glmnet (v1.9-8) (60) to first estimate a range of penalty parameters λ in 10-fold cross validation using the binomial distribution family parameter and α=0 to ensure all feature-selected genes were included in predictions. Dysplasia status was then predicted as a binary class, using lambda.min penalty.
2. wv (weighted voting) (55): Weighted voting algorithm was used to predict dysplasia status.
3. svm (Support Vector Machine) (66): The inventors used the svm function in the e1071 package (v1.6-7) (66) with linear kernel and 5-fold cross validation for class prediction.
4. rf (random forest): The randomForest package (v4.6-12) (61) was used with 1000 trees, requesting a matrix of class probabilities as output.
5. nb (Naïve Bayes): The naiveBayes function was used in the e1071 package (v1.6-7) with default parameters.
Each of the prediction algorithms generated a vector of predicted scores and a vector of predicted labels for all samples in the training and test sets.
Performance metrics. The present inventors considered 6,160 statistically and computationally viable combinations of the above parameters. The predicted class labels calculated for each model (i.e., a combination of parameters), coupled with true class labels were then used to calculate performance metrics for the biomarker as follows:
$\begin{matrix} \begin{matrix} Accuracy & \frac{T P + T N}{T P + T N + F P + F N} \\ Sensitivity & \frac{T P}{T P + F N} \\ Specificity & \frac{T N}{F P + T N} \\ Positive Predictive Value & \frac{T P}{T P + F P} \\ Nega tive Predictive Value & \frac{T N}{T N + F N} \\ Matthew' s Correlation Coefficient (MCC) & \frac{(TP \times T N) - (F P \times F N)}{\sqrt{\begin{matrix} (T P + F P) (T P + FN) \\ (T N + F P) (T N + F N) \end{matrix}}} \\ AUC for ROC \\ (Receiver Operating Characteristic) \\ MAQCII metric & 0.5 \times AUC + 0.2 5 \times (M C C + 1), \end{matrix} \\ where TP = true positives; FP = false positives; \\ TN = true negativies; FN = false negatives; \\ MCC = Matthew' s Correlation Coefficient; and \\ AUC = Area Under the Curve . \end{matrix}$
For each model, we calculated these metrics for each of the 500 iterations (different training and test sets assembled from the discovery set samples) and then averaged over all iterations. In addition to the standard performance metrics, we calculated model overfitting and gene selection consistency. The overfitting metric was calculated as the difference between the train set AUC and the test set AUC. Specifically, a model performing well on the training set but poorly on the test set would achieve a high overfitting score. For each model, the gene selection consistency metric was calculated as the average (“normalized” to biomarker size in a given model) percentage of genes passing the gene filter, that were selected into the final gene committee in all 500 iterations:
$consistency = 1 - \frac{\begin{matrix} # unique genes in all iterations - \\ biomarker size \end{matrix}}{\begin{matrix} (biomarker size \times # iterations) - \\ biomarker size \end{matrix}}$
For example, a model requiring a 10-gene biomarker would have the highest consistency (1) if it selected the same 10 genes in all 500 iterations (10 unique genes selected altogether). The same model would have the lowest consistency (0) if it selected a different set of 10 genes in all iterations (10 genes×500 iterations=5000 unique genes altogether).
Selection of best model. In selecting the best model from among the 6,160 the inventors tested and considered the degree of model overfitting, model gene selection consistency and test set AUC. First, top 10% (n=616) least overfitting models were identified. Simultaneously, the inventors identified top 10% (n=616) most consistent models. Finally, the model with the highest test set AUC among models fulfilling both criteria (n=121) was chosen as the final model.
Selection of final gene signature. The biomarker genes selected may differ between iterations due to changes in the training set. Therefore, to generate a final gene signature, the inventors trained the biomarker using all 58 discovery set samples and best model parameters.
Positive and negative controls. The biomarker discovery pipeline was also used to develop control biomarkers. As positive controls, the inventors used smoking status and sex phenotypes to identify biomarkers that could successfully distinguish former from current smokers (AUC=0.99), and females from males (AUC=0.96). As negative controls, the inventors used randomly shuffled labels for dysplasia status (AUC=0.48), smoking status (AUC=0.52), and sex (AUC=0.51). Label shuffling was conducted preserving the association between gene expression profiles and remaining phenotypes; i.e., in the case of shuffled dysplasia status, only dysplasia status was shuffled while other phenotypes and the corresponding gene expression profile remained unchanged and linked to the same sample ID.
Validations. The performance of the final biomarker was tested using the biomarker discovery pipeline in validation mode. In this mode, the pipeline takes in the entire discovery set (n=58) as the training set, and an external validation set as the test set. The test set is first corrected for gb-ratio (RNA-Seq quality metric) using limma, and the residual data is used as input. Both training and test sets are then z-score normalized. The pipeline was run using only the final model to generate prediction labels and prediction scores for the test set samples. Finally, pROC package (v1.8) (67) was used to visualize and quantify biomarker performance by plotting a ROC curve using prediction scores as the response and the dichotomous phenotype as the predictor, and extracting the AUC value from the resulting ROC object.

Detecting PML Presence in Validation Set Samples

In order to validate the biomarker's ability to detect the presence of PMLs, the performance of the biomarker was tested in smokers with and without PMLs (n=17 samples) left out of the biomarker discovery process. To assess the robustness of the results, we randomly permuted dysplasia status labels 100 times, obtaining biomarker scores for all 17 samples in each of the iterations. The present inventors then concatenated the 100 newly generated biomarker score sets for randomized labels, creating a predictor vector consisting of 1700 scores. Similarly, the inventors concatenated 100 identical copies of biomarker score sets for true labels, creating a response vector of the same length. This allowed the inventors to visualize the performance of the biomarker on true and randomized labels in a single ROC curve (FIG. 5).

Predicting PML Progression in Longitudinally-Collected Samples

In order to validate the biomarker's ability to predict sample progression/regression, the present inventors first used the biomarker to score the longitudinally collected RPCI samples (n=51). Next, calculated the difference in scores between two consecutive time points were calculated for each patient (later time point biomarker score−earlier time point biomarker score). For example, a subject with 3 samples from 3 different time points would have 3 scores, and thus two score differences could be calculated; a subject with 2 samples from 2 time points would have 2 scores, and thus 1 score difference.
Each pair of samples was assigned a “progressing/stable” or “regressing” phenotype. A “progressing/stable” phenotype indicated that the worst histological grade of PMLs sampled during the baseline procedure increased in severity or remained unchanged at follow-up; while a “regressing” phenotype indicated that the worst histological grade of PMLs sampled at baseline decreased in severity at follow-up.
The ability of the score difference to predict the “progression/regression” phenotype was quantified by plotting a ROC curve, using the vector of score differences as the predictor variable, and the progression/regression phenotype as the response variable.
Implementation of the method. The framework and structure of this pipeline are based on principles outlined for microarray data applications. The pipeline outlined in this paper was substantially modified to accommodate RNA-Seq data as well as RNA-Seq-specific methods.
Subject Inclusion/Exclusion Criteria for Samples from the British Columbia Cancer Agency (BCCA)
The samples with normal/hyperplasia histology are part of the Pan-Canadian Study and included subjects between 50 and 75 years old, current or former smokers who have smoked cigarettes for 20 years or more, and that had an estimated 3-year lung cancer risk of greater than or equal to 2%. Exclusion criteria included medical conditions, such as severe heart disease, that would jeopardize the subject's safety during participation in the study, previously diagnosed lung cancer, ex-smokers of greater than or equal to 15 years, anti-coagulant treatment, and pregnancy. The subjects with airway dysplasia were participants in three different chemoprevention studies for green tea extract (n=27 samples), sulindac (n=4 samples), and myo-inositol (n=13 samples) or from the Pan-Canadian Study described above (n=6). All samples were collected at the BCCA at baseline prior to administration of therapeutic interventions. Inclusion criteria for these chemoprevention trials can be summarized as subjects between 40 and 79 years of age, current or former smokers with at least 30 pack-years, no lung cancer history or stage 0/I curatively treated NSCLC either at least 1 year or 6 months prior to the trial (depending on trial). Exclusion criteria varied by trial but included medical conditions that would jeopardize the subject's safety during participation of the study and pregnancy. See details below:

Green Tea:

Inclusion Criteria

- Women or men age 45 to 74 years of age
- Current or former smokers who have smoked at least 30 pack-years, e.g. 1 pack per day for 30 years or more (a former smoker is defined as one who has stopped smoking for one or more years)
- ECOG performance status 0 or 1
- C-Reactive Protein >1.2 mg/L
- One or more areas of dysplasia with a surface diameter larger than 1.2 mm on autofluorescence bronchoscopy
- Willing to take Polyphenon E/placebo twice a day regularly
- Since it is unknown if Polyphenon E or EGCG will cause fetal harm when administered during pregnancy, women subjects must be postmenopausal (no menstrual periods >1 year or elevated FSH >40 mIU/ml), surgically sterile, or using birth control pill. Women of childbearing age must have normal β-HCG within 14 days to exclude pregnancy.
- Normal renal and liver function defined as serum creatinine bilirubin, AST, ALT or alkaline phosphatase levels below the upper limit of normal
- Agreeing to sign, on initial interview, informed consent forms for screening procedures (sputum cytometry analysis, fluorescence bronchoscopy, and low dose spiral thoracic CT scan). Once eligibility has been determined for the chemoprevention trial participation, agreeing to sign a study-specific treatment informed consent form.

Exclusion Criteria

- Consumption of more than 7 cups of tea a week
- Use of other natural health products containing green tea compounds
- Chronic active hepatitis/liver cirrhosis
- Severe heart disease, e.g. unstable angina, chronic congestive heart failure, use of antiarrhythmic agents
- Ongoing gastric ulcer
- Have on-going rectal bleeding
- Have a history of chronic diverticulitis and/or colitis
- Experiencing symptoms of gastritis or hemorrhoids in which medical treatment is required
- Experiencing any symptomatic gastrointestinal condition that may predispose the individual to gastrointestinal bleeding
- Acute bronchitis or pneumonia within one month
- Carcinoma in-situ or invasive cancer on bronchoscopy or abnormal spiral chest CT suspicious of lung cancer
- Known reaction to Xylocaine salbutamol, midazolam, and alfentanil
- Known allergy to green tea and/or corn starch, gelatin, or other nonmedicinal ingredients
- Any medical condition, such as acute or chronic respiratory failure, or bleeding disorder, that in the opinion of the investigator could jeopardize the subject's safety during participation in the study
- On anti-coagulant treatment such as warfarin or heparin
- Breastfeeding
- Pregnancy
- Unwilling to have a bronchoscopy
- Unwilling to have a spiral chest CT
- Unwilling to sign a consent

Sulindac:

Inclusion Criteria

- Men and women 40 through 79 years of age
- Current or former smokers with a >30 pack-year smoking history and (a) no prior lung cancer, (b) stage I NSCLC resected at least one year prior to Registration/Randomization, or (c) stage I Non-Small Cell Lung Cancer (NSCLC) with a >1 year interval since adjuvant chemotherapy conclusion
- Women of childbearing potential and men must agree to use adequate contraception (hormonal or barrier method of birth control; abstinence) prior to study entry and for the duration of study participation. Should a woman become pregnant or suspect she is pregnant while participating in this study, she should inform her treating physician immediately.
- A negative (serum or urine) pregnancy test done <7 days prior to
- Registration/Randomization, for women of childbearing potential only
- Willingness to provide tissue blocks and sputum samples for research purposes
- Participants must have normal organ and marrow function as defined below and obtained ≤45 days prior to Registration/Randomization:
- Hemoglobin ≥lower limit of institutional normal (LLN)
- Leukocytes ≥3,000/μL
- Absolute neutrophil count ≥1,500/μL
- Platelets ≥100,000/μL
- Direct bilirubin ≤1.5×institutional upper limit of normal (ULN)
- ALT (SGPT)≤1.5×institutional ULN
- Creatinine ≤1.5×institutional ULN or calculated creatinine clearance ≥30 ml/min
- ≥1 site of histologically-confirmed bronchial dysplasia
- ECOG performance status ≤1
- Negative chest x-ray
- Negative electrocardiogram

Exclusion Criteria

- Prior history of cancer (within the previous 3-years). Exception: Stage I NSCLC as outlined above, nonmelanomatous skin cancer, localized prostate cancer, carcinoma in situ (CIS) of cervix, or superficial bladder cancer with conclusion of treatment >6 months prior to Registration/Randomization.
- Prior pneumonectomy
- Solid organ transplant recipients
- History of GI ulceration, bleeding or perforation
- Uncontrolled intercurrent illness including, but not limited to: ongoing or active infection, symptomatic congestive heart failure, unstable angina pectoris, cardiac arrhythmia, recent (<6 months) history of MI, chronic renal disease, chronic liver disease, difficult to control hypertension or psychiatric illness/social situations that would limit compliance with study requirements.
- Recent (<6 months) participation in another chemoprevention trial
- Participant currently receiving any other investigational agents
- Any supplemental oxygen use (continuous or intermittent use) or documented
- Room Air (RA) SaO2<90%
- Pregnant women. Note: because there are no adequate, well-controlled studies in pregnant women and sulindac is absolutely contraindicated in the 3rd trimester.
- Breastfeeding women. Note: because there is an unknown but potential risk for adverse events in nursing infants secondary to treatment of the mother with sulindac, women who are breast-feeding will be excluded.
- Individuals who are known to be HIV positive. Note: HIV positive individuals are excluded for the following two reasons. First, HIV positive individuals are known to have altered immune function. Since one of the potential mechanisms of action of sulindac is proposed to be enhancement of immune function in preventing lung cancer progression, it is not known how the presence of HIV infection would alter this enhancement of immune function as compared to non-HIV infected individuals. Second, individuals with HIV are also known to be at higher risk for lung cancer then non-HIV infected individuals which would alter the risk/incidence of lung cancer in our study population.
- Regular NSAID or corticosteroid use during the 6-month period prior to intervention (may be eligible after washout period of 12 weeks for NSAIDs and 6 weeks for corticosteroids)
- Regular aspirin use. Exception: Aspirin can be used if prescribed by a physician for prevention. Maximum of one aspirin (81 mg) per day allowed.
- History of allergic reactions or hypersensitivity to sulindac or other NSAIDS, including aspirin-sensitive asthma
- Women of childbearing potential who are unwilling to employ adequate contraception (hormonal or barrier method of birth control; abstinence) prior to study entry and for the duration of study participation. Note: Effects of sulindac on the developing human fetus at the recommended therapeutic dose are fetal harm early in pregnancy. However, there are known harmful adverse events in the third trimester of pregnancy. Should a woman become pregnant or suspect she is pregnant while participating in this study, she should inform her treating physician immediately.
- Current use of methotrexate, corticosteroids, (anti-platelet agents) warfarin, ticlopidine, clopidogrel, aspirin, abciximab, dipyridamole, eptifibatide, tirofiban, lithium, cyclosporine, hydralazine, ACE inhibitors

Myo-Inositol:

Inclusion Criteria

- Ability to understand and willingness to sign a written informed consent document
- Age ≥45 to ≤79
- ECOG performance status (PS) 0 or 1
- One or both of the following: Stage 0/I curatively treated non-small cell lung cancer (NSCLC) with a ≥30 pack-year smoking history (surgery, adjuvant chemotherapy or radiotherapy must be completed ≥6 months prior to screening); OR Current or former smokers with a ≥30 pack-year smoking history without a history of lung cancer. Pack-years is determined by multiplying the number of packs smoked per day by the number of years smoked.
- Women of childbearing capacity who agree to use an acceptable form of birth control for the duration of the study (e.g. condom, oral contraceptives, etc.)

Exclusion Criteria

- Prior history of cancer, with the following exceptions:
- ≥3-year disease free interval (with the exception of stage I NSCLC as described above)
- Non-melanomatous skin cancer
- Localized prostate cancer with conclusion of treatment ≥6 months prior to screening
- Carcinoma in situ (CIS) of cervix with conclusion of treatment ≥6 months prior to screening
- Superficial bladder cancer with conclusion of treatment ≥6 months prior to screening
- Prior pneumonectomy
- Solid organ transplant recipients
- Uncontrolled intercurrent illness including, but not limited to: ongoing or active infection, symptomatic congestive heart failure, unstable angina pectoris, cardiac arrhythmia, severe chronic obstructive pulmonary disease requiring supplemental oxygen, difficult to control hypertension, or psychiatric illness/social situations that would limit compliance with study requirements.
- Schizophrenia
- Bipolar disorder
- Lithium treatment
- Carbamazepine treatment
- Valproate treatment
- Diabetes
- Currently using other natural health products containing inositol
- Anticoagulant use such as Coumadin or heparin. Exception: participant is off those drugs for ≥7 days prior to pre-registration.
- Recent (≤6 months) participation in another chemoprevention trial
- Participant currently receiving any other investigational agents
- Any supplemental oxygen use (continuous or intermittent use) or documented Room Air (RA) SaO₂<90%
- Pregnant women. (Excluded because the effects of high doses of myo-inositol on the fetus or newborn are not known.)
- Breastfeeding women. (Excluded because the risk for adverse events in nursing infants secondary to treatment of the mother with high doses of myo-inositol are not known.)
- History of allergic reactions attributed to myo-inositol
- History of allergies to any ingredient in the study product or placebo

Early Detection of Lung Cancer—A Pan-Canadian Study:

Inclusion Criteria

- Women or men age 50 to 75 years
- Current or former smokers who have smoked cigarettes for 20 years or more (a former smoker is defined as one who has stopped smoking for one or more years)
- An estimated 3-year lung cancer risk of >2% based on the risk prediction model.
- ECOG performance status 0 or 1
- Capable of providing, informed consent for screening procedures (low dose spiral CT, AFB, spirometry, blood biomarkers)

Exclusion Criteria

- Any medical condition, such as severe heart disease (e.g. unstable angina, chronic congestive heart failure), acute or chronic respiratory failure, bleeding disorder, that in the opinion of the investigator could jeopardize the subject's safety during participation in the study or unlikely to benefit from screening due to shortened life-expectancy from the co-morbidities
- Have been previously diagnosed with lung cancer
- Have had other cancer with the exception of the following cancers which can be included in the study: non-melanomatous skin cancer, localized prostate cancer, carcinoma in situ (CIS) of the cervix, or superficial bladder cancer. Treatment of the exceptions must have ended >6 months before registration into this study.
- Ex-smoker for ≥15 years
- On anti-coagulant treatment such as warfarin or heparin
- Known reaction to Xylocaine, salbutamol, midazolam, and alfentanil
- Pregnancy
- Unwilling to have a spiral chest CT
- Chest CT within 2 years
- Unwilling to sign a consent
  Subject Inclusion/Exclusion Criteria for Samples from RPCI

Subjects met the following high-risk lung screening criteria: 1) Personal cancer history of the lung, bronchus, head/neck, and/or esophagus and no evidence of disease at the time of enrollment, or 2.) No personal history of upper aerodigestive cancer, age 50+, and a current smoker or a former smoker with 20+ pack years. In addition, subjects in the second group had to have one or more risk factors including chronic lung disease such as emphysema, chronic bronchitis, or chronic obstructive pulmonary disease, occupationally related asbestos disease, or a family history of lung cancer in a first degree relative.

TABLE 1

Demographic and clinical characteristics stratified
by premalignant lesion status.
Data are means (SD) for continuous variables
and proportions with percentages for

	Overall	No Lesions	Lesions
Factor	(n = 82)	(n = 25)	(n = 50)	P*

Age	62.9 (7.2)	64.5 (5.8)	62.2 (8.0)	0.16
Male	54/82 (65.9)	16/25 (64)	35/50 (70)	0.61
Current smoker	40/82 (48.8)	11/25 (44)	25/50 (50)	0.81
Pack-years	47.3 (15.7)	47.6 (17.9)	47.2 (15.2)	0.93
FEV1% Predicted	82.5 (18.6)	84.5 (17.9)	81.7 (19.2)	0.54
FEV1/FVC Ratio	71.2 (7.9)	73.4 (7.4)	69.6 (8.1)	0.05
COPD (FEV1% < 80 &	24/82 (29.3)	5/25 (20)	17/50 (34)	0.28
FEV1/FVC < 70)
Histology				<10.001
Normal	12/82 (14.6)	12/25 (48)
Hyperplasia	13/82 (15.9)	13/25 (52)
Metaplasia	7/82 (8.5)
Mild Dysplasia	35/82 (42.7)		35/50 (70)
Moderate Dysplasia	12/82 (14.6)		12/50 (24)
Severe Dysplasia	3/82 (3.7)		3/50 (6)

dichotomous variables. P* values are for the comparison of subjects with and without premalignant lesions. Two sample t-tests were used for continuous variables; Fisher's exact test was used for categorical variables.

TABLE 2

Alignment statistics stratified by premalignant lesion status
Data are means (SD) for continuous variables and proportions
with percentages for dichotomous variables. Reads are
expressed in millions denoted by M. P* values are

	Overall	No Lesions	Lesions
Factor	(n = 82)	(n = 25)	(n = 50)	P*

Total Alignments	90M (17M)	90M (15M)	91M (19M)	0.78
Unique Alignments	83M (16M)	83M (14M)	84M (17M)	0.76
Properly Paired	66M (12M)	66M (11M)	67M (14M)	0.75
Alignments
Genebody
80/20 Ratio	1.3 (0.2)	1.3 (0.1)	1.3 (0.2)	0.84
Mean GC Content	47.8 (3.4)	47.4 (2.9)	48.2 (3.7)	0.34

for the comparison of subjects with and without premalignant lesions. Two sample t-tests were used for continuous variables; Fisher's exact test was used for factors.

TABLE 3

280 genes differentially expressed between subjects with PMLs and without
PMLs

Ensembl	entrezgene	hgnc_symbol	gene_biotype	wikigene_description	Direction

ENSG00000223959	172	AFG3L1P	pseudogene	AFG3 ATPase	Down-regulated in
				family gene 3-like	the presence of
				1 (S. cerevisiae),	dyplasia
				pseudogene
ENSG00000115282	64427	TTC31	protein_coding	tetratricopeptide	Down-regulated in
				repeat domain 31	the presence of
					dyplasia
ENSG00000139631	51380	CSAD	protein_coding	cysteine sulfinic	Down-regulated in
				acid decarboxylase	the presence of
					dyplasia
ENSG00000198198	23334	SZT2	protein_coding	seizure threshold 2	Down-regulated in
				homolog (mouse)	the presence of
					dyplasia
ENSG00000167524	124923		protein_coding	uncharacterized	Down-regulated in
				serine/threonine-	the presence of
				protein kinase	dyplasia
				SgK494
ENSG00000242028	25764	Cl5orf63	protein_coding	chromosome 15	Down-regulated in
				open reading frame	the presence of
				63	dyplasia
ENSG00000235194	NA	PPP1R3E	protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000179979	285464	CRIPAK	protein_coding	cysteine-rich PAK1	Down-regulated in
				inhibitor	the presence of
					dyplasia
ENSG00000164970	203259	FAM219A	protein_coding	family with	Down-regulated in
				sequence similarity	the presence of
				219, member A	dyplasia
ENSG00000162231	10482	NXF1	protein_coding	nuclear RNA	Down-regulated in
				export factor 1	the presence of
					dyplasia
ENSG00000010322	11188	NISCH	protein_coding	nischarin	Down-regulated in
					the presence of
					dyplasia
ENSG00000121310	55268	ECHDC2	protein_coding	enoyl CoA	Down-regulated in
				hydratase domain	the presence of
				containing 2	dyplasia
ENSG00000167978	23524	SRRM2	protein_coding	serine/arginine	Down-regulated in
				repetitive matrix 2	the presence of
					dyplasia
ENSG00000229180	NA		lincRNA		Down-regulated in
					the presence of
					dyplasia
ENSG00000108799	2145	EZH1	protein_coding	enhancer of zeste	Down-regulated in
				homolog 1	the presence of
				(Drosophila)	dyplasia
ENSG00000070476	79364	ZXDC	protein_coding	ZXD family zinc	Down-regulated in
				finger C	the presence of
					dyplasia
ENSG00000186088	54103	PION	protein_coding	pigeon homolog	Down-regulated in
				(Drosophila)	the presence of
					dyplasia
ENSG00000132680	22889	KIAA0907	protein_coding	KIAA0907	Down-regulated in
					the presence of
					dyplasia
ENSG00000122965	9904	RBM19	protein_coding	RNA binding motif	Down-regulated in
				protein 19	the presence of
					dyplasia
ENSG00000130766	83667	SESN2	protein_coding	sestrin 2	Down-regulated in
					the presence of
					dyplasia
ENSG00000064607	10147	SUGP2	protein_coding	SURP and G patch	Down-regulated in
				domain containing	the presence of
				2	dyplasia
ENSG00000184863	155435	RBM33	protein_coding	RNA binding motif	Down-regulated in
				protein 33	the presence of
					dyplasia
ENSG00000214021	26140	TTLL3	protein_coding	tubulin tyrosine	Down-regulated in
				ligase-like family,	the presence of
				member 3	dyplasia
ENSG00000080603	10847	SRCAP	protein_coding	Snf2-related	Down-regulated in
				CREBBP activator	the presence of
				protein	dyplasia
ENSG00000072121	23503	ZFYVE26	protein_coding	zinc finger, FYVE	Down-regulated in
				domain containing	the presence of
				26	dyplasia
ENSG00000182873	NA		antisense		Down-regulated in
					the presence of
					dyplasia
ENSG00000104365	3551	IKBKB	protein_coding	inhibitor of kappa	Down-regulated in
				light polypeptide	the presence of
				gene enhancer in	dyplasia
				B-cells, kinase beta
ENSG00000167522	29123	ANKRD11	protein_coding	ankyrin repeat	Down-regulated in
				domain 11	the presence of
					dyplasia
ENSG00000213190	10962	MLLT11	protein_coding	myeloid/lymphoid	Down-regulated in
				or mixed-lineage	the presence of
				leukemia (trithorax	dyplasia
				homolog,
				Drosophila);
				translocated to, 11
ENSG00000135407	10677	AVIL	protein_coding	advillin	Down-regulated in
					the presence of
					dyplasia
ENSG00000185219	353274	ZNF445	protein_coding	zinc finger protein	Down-regulated in
				445	the presence of
					dyplasia
ENSG00000163486	23380	SRGAP2	protein_coding	SLIT-ROBO Rho	Down-regulated in
				GTPase activating	the presence of
				protein 2	dyplasia
ENSG00000087266	6452	SH3BP2	protein_coding	SH3-domain	Down-regulated in
				binding protein 2	the presence of
					dyplasia
ENSG00000198563	692199	DDX39B	protein_coding	DEAD (Asp-Glu-	Down-regulated in
				Ala-Asp) box	the presence of
				polypeptide 39B	dyplasia
ENSG00000142528	25888	ZNF473	protein_coding	zinc finger protein	Down-regulated in
				473	the presence of
					dyplasia
ENSG00000123064	79039	DDX54	protein_coding	DEAD (Asp-Glu-	Down-regulated in
				Ala-Asp) box	the presence of
				polypeptide 54	dyplasia
ENSG00000042062	140876	FAM65C	protein_coding	family with	Down-regulated in
				sequence similarity	the presence of
				65, member C	dyplasia
ENSG00000247484	NA	NA	NA	NA	Down-regulated in
					the presence of
					dyplasia
ENSG00000100201	10521	DDX17	protein_coding	DEAD (Asp-Glu-	Down-regulated in
				Ala-Asp) box	the presence of
				helicase 17	dyplasia
ENSG00000125633	54520	CCDC93	protein_coding	coiled-coil domain	Down-regulated in
				containing 93	the presence of
					dyplasia
ENSG00000257479	NA		lincRNA		Down-regulated in
					the presence of
					dyplasia
ENSG00000076108	11176	BAZ2A	protein_coding	bromodomain	Down-regulated in
				adjacent to zinc	the presence of
				finger domain, 2A	dyplasia
ENSG00000137221	93643	TJAP1	protein_coding	tight junction	Down-regulated in
				associated protein 1	the presence of
				(peripheral)	dyplasia
ENSG00000215424	114044	MCM3	lincRNA	MCM3AP	Down-regulated in
		AP-AS1		antisense RNA 1	the presence of
				(non-protein	dyplasia
				coding)
ENSG00000100941	5411	PNN	protein_coding	pinin, desmosome	Down-regulated in
				associated protein	the presence of
					dyplasia
ENSG00000170949	90338	ZNF160	protein_coding	zinc finger protein	Down-regulated in
				160	the presence of
					dyplasia
ENSG00000240053	58496	LY6G5B	protein_coding	lymphocyte antigen	Down-regulated in
				6 complex, locus	the presence of
				G5B	dyplasia
ENSG00000181523	6448	SGSH	protein_coding	N-	Down-regulated in
				sulfoglucosamine	the presence of
				sulfohydrolase	dyplasia
ENSG00000131398	3748	KCNC3	protein_coding	potassium voltage-	Down-regulated in
				gated channel,	the presence of
				Shaw-related	dyplasia
				subfamily, member
				3
ENSG00000129933	23383	MAU2	protein_coding	MAU2 chromatid	Down-regulated in
				cohesion factor	the presence of
				homolog (C.	dyplasia
				elegans)
ENSG00000161010	51149	C5orf45	protein_coding	chromosome 5	Down-regulated in
				open reading frame	the presence of
				45	dyplasia
ENSG00000110888	65981	CAPRIN2	protein_coding	caprin family	Down-regulated in
				member 2	the presence of
					dyplasia
ENSG00000130254	9667	SAFB2	protein_coding	scaffold attachment	Down-regulated in
				factor B2	the presence of
					dyplasia
ENSG00000184634	9968	MED12	protein_coding	mediator complex	Down-regulated in
				subunit 12	the presence of
					dyplasia
ENSG00000077157	4660	PPP1R12B	protein_coding	protein phosphatase	Down-regulated in
				1, regulatory	the presence of
				subunit 12B	dyplasia
ENSG00000133624	79970	ZNF767	pseudogene	zinc finger family	Down-regulated in
				member 767	the presence of
					dyplasia
ENSG00000227372	57212	TP73-AS1	lincRNA	TP73 antisense	Down-regulated in
				RNA 1 (non-	the presence of
				protein coding)	dyplasia
ENSG00000100813	22985	ACIN1	protein_coding	apoptotic	Down-regulated in
				chromatin	the presence of
				condensation	dyplasia
				inducer 1
ENSG00000127511	23309	SIN3B	protein_coding	SIN3 transcription	Down-regulated in
				regulator homolog	the presence of
				B (yeast)	dyplasia
ENSG00000155363	4343	MOV10	protein_coding	Mov10, Moloney	Down-regulated in
				leukemia virus 10,	the presence of
				homolog (mouse)	dyplasia
ENSG00000124222	8675	STX16	protein_coding	syntaxin 16	Down-regulated in
					the presence of
					dyplasia
ENSG00000099331	4650	MYO9B	protein_coding	myosin IXB	Down-regulated in
					the presence of
					dyplasia
ENSG00000169246	NA	NPIPL3	protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000137343	79969	ATAT1	protein_coding	alpha tubulin	Down-regulated in
				acetyltransferase 1	the presence of
					dyplasia
ENSG00000169045	3187	HNRNPH1	protein_coding	heterogeneous	Down-regulated in
				nuclear	the presence of
				ribonucleoprotein	dyplasia
				H1 (H)
ENSG00000205047	NA		protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000198853	9853	RUSC2	protein_coding	RUN and SH3	Down-regulated in
				domain containing	the presence of
				2	dyplasia
ENSG00000197375	6584	SLC22A5	protein_coding	solute carrier	Down-regulated in
				family 22 (organic	the presence of
				cation/carnitine	dyplasia
				transporter),
				member 5
ENSG00000182796	440104	TMEM198B	pseudogene	transmembrane	Down-regulated in
				protein 198B,	the presence of
				pseudogene	dyplasia
ENSG00000182944	2130	EWSR1	protein_coding	Ewing sarcoma	Down-regulated in
				breakpoint region 1	the presence of
					dyplasia
ENSG00000065526	23013	SPEN	protein_coding	spen homolog,	Down-regulated in
				transcriptional	the presence of
				regulator	dyplasia
				(Drosophila)
ENSG00000137337	9656	MDC1	protein_coding	mediator of DNA-	Down-regulated in
				damage checkpoint	the presence of
				1	dyplasia
ENSG00000186174	283149	BCL9L	protein_coding	B-cell	Down-regulated in
				CLL/lymphoma 9-	the presence of
				like	dyplasia
ENSG00000075568	23505	TMEM131	protein_coding	transmembrane	Down-regulated in
				protein 131	the presence of
					dyplasia
ENSG00000170322	4798	NFRKB	protein_coding	nuclear factor	Down-regulated in
				related to kappaB	the presence of
				binding protein	dyplasia
ENSG00000171456	171023	ASXL1	protein_coding	additional sex	Down-regulated in
				combs like 1	the presence of
				(Drosophila)	dyplasia
ENSG00000044446	5256	PHKA2	protein_coding	phosphorylase	Down-regulated in
				kinase, alpha 2	the presence of
				(liver)	dyplasia
ENSG00000166436	9866	TRIM66	protein_coding	tripartite motif	Down-regulated in
				containing 66	the presence of
					dyplasia
ENSG00000255847	NA		antisense		Down-regulated in
					the presence of
					dyplasia
ENSG00000245149	100507018		lincRNA	uncharacterized	Down-regulated in
				LOC100507018	the presence of
					dyplasia
ENSG00000253200	NA		protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000100226	9567	GTPBP1	protein_coding	GTP binding	Down-regulated in
				protein 1	the presence of
					dyplasia
ENSG00000146828	56996	SLC12A9	protein_coding	solute carrier	Down-regulated in
				family 12	the presence of
				(potassium/chloride	dyplasia
				transporters),
				member 9
ENSG00000215769	NA		protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000168297	54899	PXK	protein_coding	PX domain	Down-regulated in
				containing	the presence of
				serine/threonine	dyplasia
				kinase
ENSG00000225828	100128071		protein_coding	uncharacterized	Down-regulated in
				LOC100128071	the presence of
					dyplasia
ENSG00000115459	84173	ELMOD3	protein_coding	ELMO/CED-12	Down-regulated in
				domain containing	the presence of
				3	dyplasia
ENSG00000224660	100505696		lincRNA	uncharacterized	Down-regulated in
				LOC100505696	the presence of
					dyplasia
ENSG00000090905	27327	TNRC6A	protein_coding	trinucleotide repeat	Down-regulated in
				containing 6A	the presence of
					dyplasia
ENSG00000205885	283314		antisense	uncharacterized	Down-regulated in
				LOC283314	the presence of
					dyplasia
ENSG00000117616	57035	Clorf63	protein_coding	chromosome 1	Down-regulated in
				open reading frame	the presence of
				63	dyplasia
ENSG00000114841	25981	DNAH1	protein_coding	dynein, axonemal,	Down-regulated in
				heavy chain 1	the presence of
					dyplasia
ENSG00000132382	10514	MYBBP1A	protein_coding	MYB binding	Down-regulated in
				protein (P160) 1a	the presence of
					dyplasia
ENSG00000061936	6433	SFSWAP	protein_coding	splicing factor,	Down-regulated in
				suppressor of	the presence of
				white-apricot	dyplasia
				homolog
				(Drosophila)
ENSG00000168763	26505	CNNM3	protein_coding	cyclin M3	Down-regulated in
					the presence of
					dyplasia
ENSG00000214765	641977	SEPT7P2	pseudogene	septin 7	Down-regulated in
				pseudogene 2	the presence of
					dyplasia
ENSG00000119321	23307	FKBP15	protein_coding	FK506 binding	Down-regulated in
				protein 15, 133 kDa	the presence of
					dyplasia
ENSG00000047056	22884	WDR37	protein_coding	WD repeat domain	Down-regulated in
				37	the presence of
					dyplasia
ENSG00000165699	7248	TSC1	protein_coding	tuberous sclerosis 1	Down-regulated in
					the presence of
					dyplasia
ENSG00000168970	100137047	JMJD7-	protein_coding	JMJD7-PLA2G4B	Down-regulated in
		PLA2G4B		readthrough	the presence of
					dyplasia
ENSG00000079277	8569	MKNK1	protein_coding	MAP kinase	Down-regulated in
				interacting	the presence of
				serine/threonine	dyplasia
				kinase 1
ENSG00000115568	7701	ZNF142	protein_coding	zinc finger protein	Down-regulated in
				142	the presence of
					dyplasia
ENSG00000167615	114823	LENG8	protein_coding	leukocyte receptor	Down-regulated in
				cluster (LRC)	the presence of
				member 8	dyplasia
ENSG00000100083	26088	GGA1	protein_coding	golgi-associated,	Down-regulated in
				gamma adaptin ear	the presence of
				containing, ARF	dyplasia
				binding protein 1
ENSG00000139436	9815	GIT2	protein_coding	G protein-coupled	Down-regulated in
				receptor kinase	the presence of
				interacting ArfGAP	dyplasia
				2
ENSG00000168066	7536	SF1	protein_coding	splicing factor 1	Down-regulated in
					the presence of
					dyplasia
ENSG00000099917	51586	MED15	protein_coding	mediator complex	Down-regulated in
				subunit 15	the presence of
					dyplasia
ENSG00000091831	2099	ESR1	protein_coding	estrogen receptor 1	Down-regulated in
					the presence of
					dyplasia
ENSG00000234420	100129482	ZNF37BP	pseudogene	zinc finger protein	Down-regulated in
				37B, pseudogene	the presence of
					dyplasia
ENSG00000178971	80169	CTC1	protein_coding	CTS telomere	Down-regulated in
				maintenance	the presence of
				complex	dyplasia
				component 1
ENSG00000114982	55683	KANSL3	protein_coding	KAT8 regulatory	Down-regulated in
				NSL complex	the presence of
				subunit 3	dyplasia
ENSG00000148840	23082	PPRC1	protein_coding	peroxisome	Down-regulated in
				proliferator-	the presence of
				activated receptor	dyplasia
				gamma,
				coactivator-related
				1
ENSG00000112941	11044	PAPD7	protein_coding	PAP associated	Down-regulated in
				domain containing	the presence of
				7	dyplasia
ENSG00000143624	65123	INTS3	protein_coding	integrator complex	Down-regulated in
				subunit 3	the presence of
					dyplasia
ENSG00000139990	8816	DCAF5	protein_coding	DDB1 and CUL4	Down-regulated in
				associated factor 5	the presence of
					dyplasia
ENSG00000100650	6430	SRSF5	protein_coding	serine/arginine-rich	Down-regulated in
				splicing factor 5	the presence of
					dyplasia
ENSG00000133460	66035	SLC2A11	protein_coding	solute carrier	Down-regulated in
				family 2 (facilitated	the presence of
				glucose	dyplasia
				transporter),
				member 11
ENSG00000102125	6901	TAZ	protein_coding	tafazzin	Down-regulated in
					the presence of
					dyplasia
ENSG00000136828	9649	RALGPS1	protein_coding	Ral GEF with PH	Down-regulated in
				domain and SH3	the presence of
				binding motif 1	dyplasia
ENSG00000235027	NA		antisense		Down-regulated in
					the presence of
					dyplasia
ENSG00000235706	400242	DICER1-	lincRNA	DICER1 antisense	Down-regulated in
		AS1		RNA 1 (non-	the presence of
				protein coding)	dyplasia
ENSG00000205890	100128770		antisense	uncharacterized	Down-regulated in
				LOC100128770	the presence of
					dyplasia
ENSG00000133943	80017	C14orf159	protein_coding	chromosome 14	Down-regulated in
				open reading frame	the presence of
				159	dyplasia
ENSG00000100068	91355	LRP5L	protein_coding	low density	Down-regulated in
				lipoprotein	the presence of
				receptor-related	dyplasia
				protein 5-like
ENSG00000234616	NA	JRK	processed_		Down-regulated in
			transcript		the presence of
					dyplasia
ENSG00000115687	23178	PASK	protein_coding	PAS domain	Down-regulated in
				containing	the presence of
				serine/threonine	dyplasia
				kinase
ENSG00000243335	154881	KCTD7	protein_coding	RAB guanine	Down-regulated in
				nucleotide	the presence of
				exchange factor	dyplasia
				(GEF) 1
ENSG00000131149	23199	KIAA0182	protein_coding	KIAA0182	Down-regulated in
					the presence of
					dyplasia
ENSG00000184677	9923	ZBTB40	protein_coding	zinc finger and	Down-regulated in
				BTB domain	the presence of
				containing 40	dyplasia
ENSG00000116580	54856	GON4L	protein_coding	gon-4-like (C.	Down-regulated in
				elegans)	the presence of
					dyplasia
ENSG00000130684	26152	ZNF337	protein_coding	zinc finger protein	Down-regulated in
				337	the presence of
					dyplasia
ENSG00000143442	23126	POGZ	protein_coding	pogo transposable	Down-regulated in
				element with ZNF	the presence of
				domain	dyplasia
ENSG00000249093	NA	NA	NA	NA	Down-regulated in
					the presence of
					dyplasia
ENSG00000173064	283450	C12orf51	protein_coding	chromosome 12	Down-regulated in
				open reading frame	the presence of
				51	dyplasia
ENSG00000215039	678655		lincRNA	uncharacterized	Down-regulated in
				LOC678655	the presence of
					dyplasia
ENSG00000178038	259173	ALS2CL	protein_coding	ALS2 C-terminal	Down-regulated in
				like	the presence of
					dyplasia
ENSG00000258461	NA		processed_		Down-regulated in
			transcript		the presence of
					dyplasia
ENSG00000146830	64599	GIGYF1	protein_coding	GRB10 interacting	Down-regulated in
				GYF protein 1	the presence of
					dyplasia
ENSG00000234290	NA		antisense		Down-regulated in
					the presence of
					dyplasia
ENSG00000120318	64411	ARAP3	protein_coding	ArfGAP with	Down-regulated in
				RhoGAP domain,	the presence of
				ankyrin repeat and	dyplasia
				PH domain 3
ENSG00000162241	283130	SLC25A45	protein_coding	solute carrier	Down-regulated in
				family 25, member	the presence of
				45	dyplasia
ENSG00000205268	5150	PDE7A	protein_coding	phosphodiesterase	Down-regulated in
				7A	the presence of
					dyplasia
ENSG00000160712	3570	IL6R	protein_coding	interleukin 6	Down-regulated in
				receptor	the presence of
					dyplasia
ENSG00000119906	55719	FAM178A	protein_coding	family with	Down-regulated in
				sequence similarity	the presence of
				178, member A	dyplasia
ENSG00000166762	117155	CATSPER2	protein_coding	cation channel,	Down-regulated in
				sperm associated 2	the presence of
					dyplasia
ENSG00000203709	NA	Clorf132	protein_coding		Down-regulated in
					the presence of
					dyplasia
ENSG00000167202	23102	TBC1D2B	protein_coding	TBC1 domain	Down-regulated in
				family, member 2B	the presence of
					dyplasia
ENSG00000140326	146059	CDAN1	protein_coding	congenital	Down-regulated in
				dyserythropoietic	the presence of
				anemia, type I	dyplasia
ENSG00000238105	55592		pseudogene	golgin A2	Down-regulated in
				pseudogene 5	the presence of
					dyplasia
ENSG00000167395	9726	ZNF646	protein_coding	zinc finger protein	Down-regulated in
				646	the presence of
					dyplasia
ENSG00000109063	4621	MYH3	protein_coding	myosin, heavy	Down-regulated in
				chain 3, skeletal	the presence of
				muscle, embryonic	dyplasia
ENSG00000196689	7442	TRPV1	protein_coding	transient receptor	Down-regulated in
				potential cation	the presence of
				channel, subfamily	dyplasia
				V, member 1
ENSG00000168488	11273	ATXN2L	protein_coding	ataxin 2-like	Down-regulated in
					the presence of
					dyplasia
ENSG00000230124	100527964		antisense	uncharacterized	Down-regulated in
				LOC100527964	the presence of
					dyplasia
ENSG00000184551	NA		pseudogene		Down-regulated in
					the presence of
					dyplasia
ENSG00000198026	63925	ZNF335	protein_coding	zinc finger protein	Down-regulated in
				335	the presence of
					dyplasia
ENSG00000166887	23339	VPS39	protein_coding	vacuolar protein	Down-regulated in
				sorting 39 homolog	the presence of
				(S. cerevisiae)	dyplasia
ENSG00000006530	55750	AGK	protein_coding	acylglycerol kinase	Down-regulated in
					the presence of
					dyplasia
ENSG00000128191	100302197	DGCR8	protein_coding	DiGeorge	Down-regulated in
				syndrome critical	the presence of
				region gene 8	dyplasia
ENSG00000109118	57649	PHF12	protein_coding	PHD finger protein	Down-regulated in
				12	the presence of
					dyplasia
ENSG00000068400	56850	GRIPAP1	protein_coding	GRIP1 associated	Down-regulated in
				protein 1	the presence of
					dyplasia
ENSG00000228544	100131193		antisense	uncharacterized	Down-regulated in
				LOC100131193	the presence of
					dyplasia
ENSG00000204842	6311	ATXN2	protein_coding	ataxin 2	Down-regulated in
					the presence of
					dyplasia
ENSG00000084774	790	CAD	protein_coding	carbamoyl-	Down-regulated in
				phosphate	the presence of
				synthetase 2,	dyplasia
				aspartate
				transcarbamylase,
				and dihydroorotase
ENSG00000184787	7327	UBE2G2	protein_coding	ubiquitin-	Down-regulated in
				conjugating	the presence of
				enzyme E2G 2	dyplasia
ENSG00000173120	22992	KDM2A	protein_coding	lysine (K)-specific	Down-regulated in
				demethylase 2A	the presence of
					dyplasia
ENSG00000215012	79680	C22orf29	protein_coding	chromosome 22	Down-regulated in
				open reading frame	the presence of
				29	dyplasia
ENSG00000135365	51317	PHF21A	protein_coding	PHD finger protein	Down-regulated in
				21A	the presence of
					dyplasia
ENSG00000157827	114793	FMNL2	protein_coding	formin-like 2	Down-regulated in
					the presence of
					dyplasia
ENSG00000112659	23113	CUL9	protein_coding	cullin 9	Down-regulated in
					the presence of
					dyplasia
ENSG00000108509	23125	CAMTA2	protein_coding	calmodulin binding	Down-regulated in
				transcription	the presence of
				activator 2	dyplasia
ENSG00000170919	100190939	TPT1-AS1	lincRNA	TPT1 antisense	Down-regulated in
				RNA 1 (non-	the presence of
				protein coding)	dyplasia
ENSG00000197622	56882	CDC42SE1	protein_coding	CDC42 small	Down-regulated in
				effector 1	the presence of
					dyplasia
ENSG00000100888	57680	CHD8	protein_coding	chromodomain	Down-regulated in
				helicase DNA	the presence of
				binding protein 8	dyplasia
ENSG00000213983	8906	AP1G2	protein_coding	adaptor-related	Down-regulated in
				protein complex 1,	the presence of
				gamma 2 subunit	dyplasia
ENSG00000130827	55558	PLXNA3	protein_coding	plexin A3	Down-regulated in
					the presence of
					dyplasia
ENSG00000198169	90987	ZNF251	protein_coding	zinc finger protein	Down-regulated in
				251	the presence of
					dyplasia
ENSG00000132424	25957	PNISR	protein_coding	PNN-interacting	Down-regulated in
				serine/arginine-rich	the presence of
				protein	dyplasia
ENSG00000120709	51307	FAM53C	protein_coding	family with	Down-regulated in
				sequence similarity	the presence of
				53, member C	dyplasia
ENSG00000131067	2686	GGT7	protein_coding	gamma-	Down-regulated in
				glutamyltransferase	the presence of
				7	dyplasia
ENSG00000166888	6778	STAT6	protein_coding	signal transducer	Down-regulated in
				and activator of	the presence of
				transcription 6,	dyplasia
				interleukin-4
				induced
ENSG00000258727	NA		antisense		Down-regulated in
					the presence of
					dyplasia
ENSG00000141867	23476	BRD4	protein_coding	bromodomain	Down-regulated in
				containing 4	the presence of
					dyplasia
ENSG00000005339	1387	CREBBP	protein_coding	CREB binding	Down-regulated in
				protein	the presence of
					dyplasia
ENSG00000165275	158234	RG9MTD3	protein_coding	RNA (guanine-9-)	Down-regulated in
				methyltransferase	the presence of
				domain containing	dyplasia
				3
ENSG00000196535	399687	MYO18A	protein_coding	myosin XVIIIA	Down-regulated in
					the presence of
					dyplasia
ENSG00000125814	63908	NAPB	protein_coding	N-ethylmaleimide-	Down-regulated in
				sensitive factor	the presence of
				attachment protein,	dyplasia
				beta
ENSG00000092421	57556	SEMA6A	protein_coding	sema domain,	Down-regulated in
				transmembrane	the presence of
				domain (TM), and	dyplasia
				cytoplasmic
				domain,
				(semaphorin) 6A
ENSG00000137497	4926	NUMA1	protein_coding	nuclear mitotic	Down-regulated in
				apparatus protein 1	the presence of
					dyplasia
ENSG00000100416	55687	TRMU	protein_coding	tRNA 5-	Down-regulated in
				methylaminomethy	the presence of
				1-2-thiouridylate	dyplasia
				methyltransferase
ENSG00000110274	22897	CEP164	protein_coding	centrosomal protein	Down-regulated in
				164 kDa	the presence of
					dyplasia
ENSG00000104885	84444	DOT1L	protein_coding	DOT1-like, histone	Down-regulated in
				H3	the presence of
				methyltransferase	dyplasia
				(S. cerevisiae)
ENSG00000244161	100506906	FLNB-	antisense	FLNB antisense	Down-regulated in
		AS1		RNA 1 (non-	the presence of
				protein coding)	dyplasia
ENSG00000218418	NA		pseudogene		Down-regulated in
					the presence of
					dyplasia
ENSG00000171163	55657	ZNF692	protein_coding	zinc finger protein	Down-regulated in
				692	the presence of
					dyplasia
ENSG00000184313	374977	HEATR8	protein_coding	HEAT repeat	Down-regulated in
				containing 8	the presence of
					dyplasia
ENSG00000156858	78994	PRR14	protein_coding	proline rich 14	Down-regulated in
					the presence of
					dyplasia
ENSG00000247743	NA	NA	NA	NA	Down-regulated in
					the presence of
					dyplasia
ENSG00000213015	51157	ZNF580	protein_coding	zinc finger protein	Down-regulated in
				580	the presence of
					dyplasia
ENSG00000142937	94163	RPS8	protein_coding	ribosomal protein	Up-regulated in the
				S8	presence of
					dyplasia
ENSG00000129518	55837	EAPP	protein_coding	E2F-associated	Up-regulated in the
				phosphoprotein	presence of
					dyplasia
ENSG00000213326	NA	RPS7P11	pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000177889	7334	UBE2N	protein_coding	ubiquitin-	Up-regulated in the
				conjugating	presence of
				enzyme E2N	dyplasia
ENSG00000185834	NA	RPL12P4	pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000166171	25911	DPCD	protein_coding	deleted in primary	Up-regulated in the
				ciliary dyskinesia	presence of
				homolog (mouse)	dyplasia
ENSG00000235297	NA		pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000181163	4869	NPM1	protein_coding	nucleophosmin	Up-regulated in the
				(nucleolar	presence of
				phosphoprotein	dyplasia
				B23, numatrin)
ENSG00000177600	619565	RPLP2	protein_coding	ribosomal protein,	Up-regulated in the
				large, P2	presence of
					dyplasia
ENSG00000082515	29093	MRPL22	protein_coding	mitochondrial	Up-regulated in the
				ribosomal protein	presence of
				L22	dyplasia
ENSG00000185068	404672	GTF2H5	protein_coding	general	Up-regulated in the
				transcription factor	presence of
				IIH, polypeptide 5	dyplasia
ENSG00000134248	10542	HBXIP	protein_coding	hepatitis B virus x	Up-regulated in the
				interacting protein	presence of
					dyplasia
ENSG00000186198	123264		protein_coding	organic solute	Up-regulated in the
				transporter beta	presence of
					dyplasia
ENSG00000186132	130355	C2orf76	protein_coding	chromosome 2	Up-regulated in the
				open reading frame	presence of
				76	dyplasia
ENSG00000185641	NA		pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000168653	4725	NDUFS5	protein_coding	NADH	Up-regulated in the
				dehydrogenase	presence of
				(ubiquinone) Fe-S	dyplasia
				protein 5, 151 kDa
				(NADH-coenzyme
				Q reductase)
ENSG00000100554	51382	ATP6V1D	protein_coding	ATPase, H+	Up-regulated in the
				transporting,	presence of
				lysosomal 34 kDa,	dyplasia
				V1 subunit D
ENSG00000161016	6132	RPL8	protein_coding	ribosomal protein	Up-regulated in the
				L8	presence of
					dyplasia
ENSG00000111775	1337	COX6A1	protein_coding	cytochrome c	Up-regulated in the
				oxidase subunit VIa	presence of
				polypeptide 1	dyplasia
ENSG00000183978	28958	CCDC56	protein_coding	coiled-coil domain	Up-regulated in the
				containing 56	presence of
					dyplasia
ENSG00000236552	728658	RPL13AP5	pseudogene	ribosomal protein	Up-regulated in the
				L13a pseudogene 5	presence of
					dyplasia
ENSG00000236801	NA		pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000131100	529	ATP6V1E1	protein_coding	ATPase, H+	Up-regulated in the
				transporting,	presence of
				lysosomal 31 kDa,	dyplasia
				V1 subunit E1
ENSG00000235174	NA	RPL39P3	pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000169740	7580	ZNF32	protein_coding	zinc finger protein	Up-regulated in the
				32	presence of
					dyplasia
ENSG00000129562	1603	DAD1	protein_coding	defender against	Up-regulated in the
				cell death 1	presence of
					dyplasia
ENSG00000144713	6161	RPL32	protein_coding	ribosomal protein	Up-regulated in the
				L32	presence of
					dyplasia
ENSG00000197756	6168	RPL37A	protein_coding	ribosomal protein	Up-regulated in the
				L37a	presence of
					dyplasia
ENSG00000164751	5828	PEX2	protein_coding	peroxisomal	Up-regulated in the
				biogenesis factor 2	presence of
					dyplasia
ENSG00000010278	100652804	CD9	protein_coding	CD9 molecule	Up-regulated in the
					presence of
					dyplasia
ENSG00000140988	26784	RPS2	protein_coding	ribosomal protein	Up-regulated in the
				S2	presence of
					dyplasia
ENSG00000198618	NA	PPIAP22	pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000151465	8872	CDC123	protein_coding	cell division cycle	Up-regulated in the
				123 homolog (S.	presence of
				cerevisiae)	dyplasia
ENSG00000143543	10899	JTB	protein_coding	jumping	Up-regulated in the
				translocation	presence of
				breakpoint	dyplasia
ENSG00000244398	NA		pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000232856	NA		protein_coding		Up-regulated in the
					presence of
					dyplasia
ENSG00000108100	219771	CCNY	protein_coding	cyclin Y	Up-regulated in the
					presence of
					dyplasia
ENSG00000118939	7347	UCHL3	protein_coding	ubiquitin carboxyl-	Up-regulated in the
				terminal esterase	presence of
				L3 (ubiquitin	dyplasia
				thiolesterase)
ENSG00000169021	7386	UQCRFS1	protein_coding	ubiquinol-	Up-regulated in the
				cytochrome c	presence of
				reductase, Rieske	dyplasia
				iron-sulfur
				polypeptide 1
ENSG00000172809	6169	RPL38	protein_coding	ribosomal protein	Up-regulated in the
				L38	presence of
					dyplasia
ENSG00000137154	6194	RPS6	protein_coding	ribosomal protein	Up-regulated in the
				S6	presence of
					dyplasia
ENSG00000164405	27089	UQCRQ	protein_coding	ubiquinol-	Up-regulated in the
				cytochrome c	presence of
				reductase, complex	dyplasia
				III subunit VII,
				9.5 kDa
ENSG00000143457	55204	GOLPH3L	protein_coding	golgi	Up-regulated in the
				phosphoprotein 3-	presence of
				like	dyplasia
ENSG00000138297	100287932	TIMM23	protein_coding	translocase of inner	Up-regulated in the
				mitochondrial	presence of
				membrane 23	dyplasia
				homolog (yeast)
ENSG00000228474	100128731	OST4	protein_coding	oligosaccharyltrans	Up-regulated in the
				ferase 4 homolog	presence of
				(S. cerevisiae)	dyplasia
ENSG00000112981	8382	NME5	protein_coding	non-metastatic cells	Up-regulated in the
				5, protein	presence of
				expressed in	dyplasia
				(nucleoside-
				diphosphate kinase)
ENSG00000112667	10591	C6orf108	protein_coding	chromosome 6	Up-regulated in the
				open reading frame	presence of
				108	dyplasia
ENSG00000183617	116541	MRPL54	protein_coding	mitochondrial	Up-regulated in the
				ribosomal protein	presence of
				L54	dyplasia
ENSG00000188873	NA	RPL10AP2	pseudogene		Up-regulated in the
					presence of
					dyplasia
ENSG00000131143	1327	COX411	protein_coding	cytochrome c	Up-regulated in the
				oxidase subunit IV	presence of
				isoform 1	dyplasia
ENSG00000178741	9377	COX5A	protein_coding	cytochrome c	Up-regulated in the
				oxidase subunit Va	presence of
					dyplasia
ENSG00000232112	51372	CCDC72	protein_coding	coiled-coil domain	Up-regulated in the
				containing 72	presence of
					dyplasia
ENSG00000178449	84987	COX14	protein_coding	COX14	Up-regulated in the
				cytochrome c	presence of
				oxidase assembly	dyplasia
				homolog (S.
				cerevisiae)
ENSG00000138663	51138	COPS4	protein_coding	COP9 constitutive	Up-regulated in the
				photomorphogenic	presence of
				homolog subunit 4	dyplasia
				(Arabidopsis)
ENSG00000149547	9538	E124	protein_coding	etoposide induced	Up-regulated in the
				2.4 mRNA	presence of
					dyplasia
ENSG00000173660	440567	UQCRH	protein_coding	ubiquinol-	Up-regulated in the
				cytochrome c	presence of
				reductase hinge	dyplasia
				protein
ENSG00000125356	4694	NDUFA1	protein_coding	NADH	Up-regulated in the
				dehydrogenase	presence of
				(ubiquinone) 1	dyplasia
				alpha subcomplex,
				1, 7.5 kDa
ENSG00000162244	6159	RPL29	protein_coding	ribosomal protein	Up-regulated in the
				L29	presence of
					dyplasia
ENSG00000174444	595097	RPL4	protein_coding	ribosomal protein	Up-regulated in the
				L4	presence of
					dyplasia
ENSG00000145247	132299	OCIAD2	protein_coding	OCIA domain	Up-regulated in the
				containing 2	presence of
					dyplasia
ENSG00000178980	6415	SEPW1	protein_coding	selenoprotein W, 1	Up-regulated in the
					presence of
					dyplasia
ENSG00000169020	521	ATP5I	protein_coding	ATP synthase, H+	Up-regulated in the
				transporting,	presence of
				mitochondrial Fo	dyplasia
				complex, subunit E
ENSG00000125743	6633	SNRPD2	protein_coding	small nuclear	Up-regulated in the
				ribonucleoprotein	presence of
				D2 polypeptide	dyplasia
				16.5 kDa
ENSG00000101928	56180	MOSPD1	protein_coding	motile sperm	Up-regulated in the
				domain containing	presence of
				1	dyplasia
ENSG00000151366	100532726	NDUFC2	protein_coding	NADH	Up-regulated in the
				dehydrogenase	presence of
				(ubiquinone) 1,	dyplasia
				subcomplex
				unknown, 2,
				14.5 kDa
ENSG00000171421	64979	MRPL36	protein_coding	mitochondrial	Up-regulated in the
				ribosomal protein	presence of
				L36	dyplasia
ENSG00000198755	4736	RPL10A	protein_coding	ribosomal protein	Up-regulated in the
				L10a	presence of
					dyplasia
ENSG00000232119	28985	MCTS1	protein_coding	malignant T cell	Up-regulated in the
				amplified sequence	presence of
				1	dyplasia
ENSG00000198643	131177	FAM3D	protein_coding	family with	Up-regulated in the
				sequence similarity	presence of
				3, member D	dyplasia
ENSG00000123144	79002	C19orf43	protein_coding	chromosome 19	Up-regulated in the
				open reading frame	presence of
				43	dyplasia
ENSG00000111669	7167	TPI1	protein_coding	triosephosphate	Up-regulated in the
				isomerase 1	presence of
					dyplasia
ENSG00000089063	29058	TMEM230	protein_coding	chromosome 20	Up-regulated in the
				open reading frame	presence of
				30	dyplasia
ENSG00000214026	6150	MRPL23	protein_coding	mitochondrial	Up-regulated in the
				ribosomal protein	presence of
				L23	dyplasia
ENSG00000119421	4702	NDUFA8	protein_coding	NADH	Up-regulated in the
				dehydrogenase	presence of
				(ubiquinone) 1	dyplasia
				alpha subcomplex,
				8, 19 kDa
ENSG00000135940	1329	COX5B	protein_coding	cytochrome c	Up-regulated in the
				oxidase subunit Vb	presence of
					dyplasia
ENSG00000146066	192286	HIGD2A	protein_coding	HIG1 hypoxia	Up-regulated in the
				inducible domain	presence of
				family, member 2A	dyplasia
ENSG00000170892	79042	TSEN34	protein_coding	tRNA splicing	Up-regulated in the
				endonuclease 34	presence of
				homolog (S.	dyplasia
				cerevisiae)
ENSG00000166920	84419	C15orf48	protein_coding	chromosome 15	Up-regulated in the
				open reading frame	presence of
				48	dyplasia
ENSG00000140307	2958	GTF2A2	protein_coding	general	Up-regulated in the
				transcription factor	presence of
				IIA, 2, 12 kDa	dyplasia
ENSG00000184831	79135	APOO	protein_coding	apolipoprotein O	Up-regulated in the
					presence of
					dyplasia
ENSG00000205544	254863	C17orf61	protein_coding	chromosome 17	Up-regulated in the
				open reading frame	presence of
				61	dyplasia

Supplemental Table 1. ANOVA derived p-values for the association between the

surrogate variables and demographic/phenotypic variables

Variable	SV1	SV2	SV3	SV4	SV5	SV6	SV7	SV8	SV9

Presence of	0.549	0.376	0.964	0.500	0.118	0.481	0.046	0.166	0.652
premalignant lesion
(2-level)
Smoking status	0.000	0.655	0.191	0.084	0.689	0.804	0.308	0.719	0.761
Smoking status by	0.000	0.363	0.801	0.045	0.819	0.780	0.130	0.827	0.663
Gene Expression
Sex	0.961	0.058	0.000	0.032	0.492	0.801	0.433	0.884	0.991
COPD status	0.612	0.866	0.047	0.161	0.973	0.129	0.083	0.007	0.592
Pack-years	0.398	0.293	0.523	0.576	0.845	0.399	0.875	0.428	0.178
Age	0.300	0.153	0.562	0.845	0.166	0.618	0.037	0.050	0.528
FEV1	0.050	0.391	0.046	0.009	0.123	0.150	0.171	0.028	0.691
FEV1/FVC ratio	0.023	0.670	0.172	0.056	0.491	0.107	0.028	0.011	0.708
Barcode	0.870	0.605	0.006	0.500	0.745	0.444	0.695	0.119	0.187
Lane	0.335	0.748	0.682	0.351	0.037	0.792	0.402	0.996	0.549
Batch	0.676	0.730	0.474	0.426	0.861	0.037	0.145	0.688	0.261
GC content	0.599	0.886	0.057	0.902	0.257	0.157	0.001	0.416	0.210
Genebody 80/20 ratio	0.000	0.245	0.633	0.271	0.000	0.736	0.015	0.319	0.048
(gb-ratio)
Number of Uniquely	0.302	0.154	0.726	0.948	0.055	0.120	0.036	0.163	0.586
Aligning Reads
Number of Reads	0.545	0.605	0.498	0.442	0.000	0.383	0.170	0.745	0.942
Aligning to Splice
Junctions
Z-score (sample mean	0.514	0.371	0.238	0.595	0.024	0.031	0.005	0.353	0.021
of z-score normalized
data by gene)
Relative Expression	0.814	0.615	0.996	0.740	0.918	0.887	0.214	0.274	0.111
(sample median of
ratios computed for
each gene by dividing
the expression by the
median expression)

Supplemental Table 2.

Phenotypic information about the human biopsy cell

cultures used in the bioenergetics experiments.

		Smoking	Bio-	Mito-
Histology	Gender	Status	energetics	TrackerFM

Normal	F	Current	X
Normal	M	Current	X
Normal	F	Former	X
Normal	M	Former	X
Normal	F	Current	X	X
Normal	F	Current	X	X
Moderate	M	Current	X
Dysplasia
Severe	M	Former	X
Dysplasia
Severe	M	Current	X
Dysplasia
Low grade	M	Former	X
dysplasia
Severe	M	Current	X	X
Dysplasia
Low grade	M	Former	X	X
dysplasia

Supplemental Table 3.

Phenotypic information about the human biopsies

used in the IHC experiments.

		Smoking
Stain	PtID	Status	WorstHistology_Description

Tomm-22	Pt 3	FS	0 Normal, Negative,
			Benign Mucosa
Cox-IV	Pt	3	FS	0 Normal, Negative,
			Benign Mucosa
Tomm-22	Pt 4	FS	23 Squamous Metaplasia
			(non-specific),
			Mature Metaplasia,
			Squamous Hyperplasia
Cox-IV	Pt	4	FS	23 Squamous Metaplasia
			(non-specific),
			Mature Metaplasia,
			Squamous Hyperplasia
Tomm-22	Pt 3	FS	25 Moderate Dysplasia,
			Squamous Pre-invasive
Cox-IV	Pt	3	FS	25 Moderate Dysplasia,
			Squamous Pre-invasive
Tomm-22	Pt 1	CS	27 CIS Squamous
			Carcinoma In-Situ
Cox-IV	Pt	1	CS	27 CIS Squamous
			Carcinoma In-Situ

(*CS refers to current smoker and FS to former smoker)

Supplemental Table 4.

Demographic and clinical characteristics of the British Columbia Lung Health Study stratified by premalignant lesions status

Discovery Set

Validation Set

	Overall	No Lesions	Lesions		Overall	No Lesions	Lesions
Factor	(n = 58)	(n = 20)	(n = 38)	P*	(n = 17)	(n = 5)	(n = 12)	P*

Age	62.7	(7.1)	64.1	(5.8)	61.9	(7.6)	0.24	63.9	(8.6)	66	(5.8)	63	(9.7)	0.45
Male	37/58	(63.8)	12/20	(60)	25/38	(65.8)	0.78	14/17	(82.4)	4/5	(80)	10/12	(83.3)	1
Current smoker	28/58	(48.3)	9/20	(45)	19/38	(50)	0.79	8/17	(47.1)	2/5	(40)	6/12	(50)	1
Pack-years	48.2	(16.9)	49.4	(18.9)	47.5	(15.9)	0.71	44.6	(12.9)	40.5	(11.6)	46.3	(13.5)	0.39
FEV1% Predicted	86.5	(17.7)	87.8	(16.7)	85.7	(18.5)	0.66	69.5	(16.2)	71	(17.7)	68.9	(16.3)	0.83
FEV1/FVC Ratio	72.1	(7.7)	75.1	(6.3)	70.4	(8)	0.02	67	(8.1)	66.8	(8.5)	67.1	(8.3)	0.95
COPD (FEV1 % < 80 &	11/58	(19)	2/20	(10)	9/38	(23.7)	0.3	11/17	(64.7)	3/5	(60)	8/12	(66.7)	1
FEV1/FVC < 70)
Histology							<0.001							<0.001
Normal	11/58	(19)	11/20	(55)				1/17	(5.9)	1/5	(20)
Hyperplasia	9/58	(15.5)	9/20	(45)				4/17	(23.5)	4/5	(80)
Metaplasia	0/58	(0)						0/17	(0)
Mild Dysplasia	29/58	(50)			29/38	(76.3)		6/17	(35.3)			6/12	(50)
Moderate Dysplasia	6/58	(10.3)			6/38	(15.8)		6/17	(35.3)			6/12	(50)
Severe Dysplasia	3/58	(5.2)			3/38	(7.9)						0/12	(0)

Data are means (SD) for continuous variables and proportions (%) dichotomous variables. Reads are expressed in millions denoted by M. P* values are for the comparison of subjects with and without premalignant lesions. Two sample t-tests were used for continuous variables; Fisher's exact test was used for factors.

Supplemental Table 5.

Alignment statistics of the British Columbia Lung Health Study Discovery and the Roswell Park Cancer Institute cohort

BC-LHS Discovery Set

BC-LHS Validation Set

RPCI

Overall

No Lesions

Lesions

Overall

No Lesions

Lesions

Overall

Factor	(n = 58)	(n = 20)	(n = 38)	P*	(n = 17)	(n = 5)	(n = 12)		P*	(n = 51)

Total Alignments	90M	(16M)	98M	(15M)	91M	(17M)	0.67	93M	(22M)	94M	(18M)	92M	(24M)	0.86	95M	(15M)
Unique Alignments	83M	(15M)	82M	(13M)	83M	(16M)	0.65	85M	(20M)	86M	(16M)	84M	(22M)	0.85
Properly Paired Alignments	66M	(1.2M)	65M	(11M)	67M	(12M)	0.63	68M	(16M)	69M	(13M)	67M	(17M)	0.86	65M	(9.6M)
Genebody 80/20 Ratio	1.3	(0.2)	1.3	(0.1)	1.3	(0.2)	0.39	1.3	(0.3)	1.2	(0.1)	1.4	(0.3)	0.15	1.8	(0.2)
Mean GC Content	48.1	(3.4)	47.5	(2.7)	48.4	(3.6)	0.33	47.4	(3.8)	46.9	(3.8)	47.6	(3.9)	0.74	49.2	(1.4)

Data are means (SD). Reads are expressed in millions denoted by M. P* values are for two sample t-tests for comparison of subjects with and without premalignant lesions.

Supplemental Table 6.

Demographic and clinical characteristics of the Roswell

Park Cancer Institute Cohort (n = 51 samples from n = 23 subjects)

			Progressing
Factor	Overall	Regressing	Stable	P*

No. Samples	51	34	22
No. Sample	28	17	11
Pairs
No. Patients**	23	16	10
Time between	343.8 (171.9)	350.9 (199.6)	332.8 (125.9)	0.77
Procedures
(Days)
Histological	−0.9 (1.7)	−1.9 (1.0)	0.7 (1.3)	<0.001
Grade
Change

Worst Histological Lesion Observed

Normal	5/51 (9.8)	4/34 (11.8)	2/22 (9.1)	0.038
Hyperplasia	6/51 (11.8)	5/34 (14.7)	1/22 (4.5)
Metaplasia	9/51 (17.6)	8/34 (23.5)	1/22 (4.5)
Mild Dysplasia	3/51 (5.9)	3/34 (8.8)	0 (0)
Moderate	20/51 (39.2)	9/34 (26.5)	15/22 (68.2)
Dysplasia
Severe	8/51 (15.7)	5/34 (14.7)	3/22 (13.6)
Dysplasia
Age at	58.1 (6.5)	58.4 (6.9)	57.6 (6.1)	1
Baseline
Male	13/28 (46.4)	7/17 (41.2)	6/11 (54.5)	0.7
Ever smoker at	27/28 (96.4)	17/17 (100)	10/11 (90.9)	0.39
Baseline
Pack-years at	48.1 (22)	49.8 (24.8)	45.4 (17.6)	1
Baseline

Data are means (SD) for continuous variables and proportions (%) for dichotomous variables.
P*values are for the comparison of samples, sample pairs, or patients classified as having regressing or progressing/stable PMLs. Two sample t-tests were used for continuous variables; Fisher's exact test was used for factors.
**Among the 23 patients, 3 patients had 2 sample pairs where one pair was classified as regressing and the other as progressing/stable. These patients are counted in both the regressing and progressing/stable columns.


Dataset 1.
Ensembl IDs for genes used to predict smoking status.

ENSG00000151632
ENSG00000125398
ENSG00000159228
ENSG00000109586
ENSG00000049089
ENSG00000198431
ENSG00000140961
ENSG00000117450
ENSG00000111058
ENSG00000198074
ENSG00000001084
ENSG00000168309
ENSG00000108602
ENSG00000065833
ENSG00000215182
ENSG00000079819
ENSG00000117983
ENSG00000163931
ENSG00000173376
ENSG00000197838
ENSG00000176153
ENSG00000136810
ENSG00000137642
ENSG00000134873
ENSG00000172765
ENSG00000154040
ENSG00000048707
ENSG00000123124
ENSG00000102359
ENSG00000197747
ENSG00000103222
ENSG00000103647
ENSG00000099968
ENSG00000196344
ENSG00000140939
ENSG00000167996
ENSG00000006125
ENSG00000149256
ENSG00000010404
ENSG00000023909
ENSG00000077147
ENSG00000134775
ENSG00000177156
ENSG00000123700
ENSG00000124664
ENSG00000197712
ENSG00000154822
ENSG00000086548
ENSG00000137573
ENSG00000100012
ENSG00000136205
ENSG00000138061
ENSG00000104341
ENSG00000151012
ENSG00000039537
ENSG00000181458
ENSG00000006210
ENSG00000078596
ENSG00000117394
ENSG00000106541
ENSG00000125798
ENSG00000109854
ENSG00000196139
ENSG00000162496
ENSG00000181019
ENSG00000140526
ENSG00000166670
ENSG00000198417
ENSG00000162804
ENSG00000105388
ENSG00000069764
ENSG00000108924
ENSG00000171903
ENSG00000085662
ENSG00000137648
ENSG00000125144
ENSG00000113924
ENSG00000134827
ENSG00000142655
ENSG00000139629
ENSG00000160180
ENSG00000124107
ENSG00000119514
ENSG00000227051
ENSG00000144711
ENSG00000101445
ENSG00000137337
ENSG00000114638
ENSG00000142657
ENSG00000130595
ENSG00000145147
ENSG00000087842
ENSG00000133985
ENSG00000125813

Dataset 2. Results of pathway enrichment using ROAST (FDR < 0.05).

The column “Direction” refers to pathway enrichment among genes up-regulated (Up) or down-regulated (Down) in the presence of PMLs.

Pathway	NGenes	PropDown	PropUp	Direction	PValue	FDR

REACTOME_METABOLISM_OF_PROTEINS	382	0.091623	0.544503	Up	0.002	0.0128
REACTOME_METABOLISM_OF_RNA	251	0.139442	0.494024	Up	0.002	0.0128
REACTOME_METABOLISM_OF_MRNA	206	0.131068	0.533981	Up	0.002	0.0128
KEGG_HUNTINGTONS_DISEASE	158	0.126582	0.607595	Up	0.002	0.0128
KEGG_ALZTIEIMERS_DISEASE	141	0.120567	0.631206	Up	0.002	0.0128
REACTOME_TRANSLATION	141	0.042553	0.780142	Up	0.002	0.0128
REACTOME_INFLUENZA_LIFE_CYCLE	133	0.075188	0.691729	Up	0.002	0.0128
REACTOME_TCA_CYCLE_AND_RESPIRATORY_ELECTRON_	125	0.088	0.64	Up	0.002	0.0128
TRANSPORT
KEGG_OXIDATIVE_PHOSPHORYLATION	117	0.042735	0.692308	Up	0.002	0.0128
KEGG_PARKINSONS_DISEASE	113	0.079646	0.699115	Up	0.002	0.0128
REACTOME_SRP_DEPENDENT_COTRANSLATIONAL_	105	0.019048	0.885714	Up	0.002	0.0128
PROTEIN_TARGETING_TO_MEMBRANE
REACTOME_NONSENSE_MEDIATED_DECAY_	103	0.07767	0.776699	Up	0.002	0.0128
ENHANCED_BY_THE_EXON_JUNCTION_COMPLEX
REACTOME_3_UTR_MEDIATED_TRANSLATIONAL_	102	0.029412	0.843137	Up	0.002	0.0128
REGULATION
REACTOME_SIGNALING_BY_RHO_GTPASES	93	0.387097	0.150538	Down	0.002	0.0128
REACTOME_RESPIRATORY_ELECTRON_TRANSPORT_	91	0.021978	0.758242	Up	0.002	0.0128
ATP_SYNTHESIS_BY_CHEMIOSMOTIC_CO
AT_PRODUCTION_BY_UNCOUPLING_PROTEINS_KEGG_
JAK_STAT_SIGNALING_PATHWAY	87	0.321839	0.126437	Down	0.002	0.0128
KEGG_PYRIMIDINE_METABOLISM	84	0.154762	0.380952	Up	0.002	0.0128
KEGG_RIBOSOME	83	0.012048	0.939759	Up	0.002	0.0128
REACTOME_PEPTIDE_CHAIN_ELONGATION	82	0.012195	0.939024	Up	0.002	0.0128
REACTOME_RESPIRATORY_ELECTRON_TRANSPORT	74	0.013514	0.756757	Up	0.002	0.0128
PID_HDAC_CLASSI_PATHWAY	60	0.366667	0.15	Down	0.002	0.0128
PID_MYC_REPRESSPATHWAY	55	0.381818	0.127273	Down	0.002	0.0128
REACTOME_ACTIVATION_OF_THE_MRNA_UPON_	55	0.054545	0.745455	Up	0.002	0.0128
BINDING_OF_THE_CAP_BINDING_COMPLEX_A _
SUBSEQUENT_BINDING_TO_43S
PID_AVB3_INTEGRIN_PATHWAY	53	0.320755	0.132075	Down	0.002	0.0128
KEGG_ADIPOCYTOKINE_SIGNALING_PATHWAY	51	0.411765	0.176471	Down	0.002	0.0128
REACTOME_MITOCHONDRIAL_PROTEIN_IMPORT	49	0.102041	0.530612	Up	0.002	0.0128
REACTOME_FORMATION_OF_THE_TERNARY_	47	0.042553	0.829787	Up	0.002	0.0128
COMPLEX_AND_SUBSEQUENTLY_THE_43S_COMP
KEGG_CARDIAC_MUSCLE_CONTRACTION	43	0.116279	0.55814	Up	0.002	0.0128
KEGG_LYSINE_DEGRADATION	42	0.428571	0.166667	Down	0.002	0.0128
PID_IL4_2PATHWAY	42	0.380952	0.119048	Down	0.002	0.0128
REACTOME_FORMATION_OF_RNA_POL_II_	41	0.170732	0.439024	Up	0.002	0.0128
ELONGATION_COMPLEX_
KEGG_NOTCH_SIGNALING_PATHWAY
	40	0.425	0.125	Down	0.002	0.0128
PID_RHOA_REG_PATHWAY	40	0.475	0.125	Down	0.002	0.0128
REACTOME_NRAGE_SIGNALS_DEATH_THROUGH_INK	39	0.358974	0.153846	Down	0.002	0.0128
REACTOME_PRE_NOTCH_EXPRESSION_AND_	38	0.342105	0.131579	Down	0.002	0.0128
PROCESSING
REACTOME_NCAM_SIGNALING_FOR_NEURITE_OUT_	37	0.459459	0.108108	Down	0.002	0.0128
GROWTH
ST_GA13_PATHWAY	33	0.424242	0.121212	Down	0.002	0.0128
PID_RAC1_REG_PATHWAY	33	0.454545	0.121212	Down	0.002	0.0128
REACTOME_BMAL1_CLOCK_NPAS2_ACTIVATES_	33	0.484848	0.090909	Down	0.002	0.0128
CIRCADIAN_EXPRESSION
BIOCARTA_CARM_ER_PATHWAY	32	0.34375	0.125	Down	0.002	0.0128
REACTOME_G1_PHASE	32	0.09375	0.5	Up	0.002	0.0128
REACTOME_FORMATION_OF_THE_HIV1_EARLY_	31	0.129032	0.483871	Up	0.002	0.0128
ELONGATION_COMPLEX
KEGG_PROPANOATE_METABOLISM	30	0.1	0.433333	Up	0.002	0.0128
PID_FRA_PATHWAY	28	0.428571	0.071429	Down	0.002	0.0128
REACTOME_PURINE_METABOLISM	28	0.178571	0.392857	Up	0.002	0.0128
KEGG_BUTANOATE_METABOLISM	27	0.037037	0.481481	Up	0.002	0.0128
BIOCARTA_MYOSIN_PATHWAY	27	0.296296	0.111111	Down	0.002	0.0128
REACTOME_MRNA_CAPPING	27	0.111111	0.481481	Up	0.002	0.0128
REACTOME_FORMATION_OF_TRANSCRIPTION_	27	0.074074	0.518519	Up	0.002	0.0128
COUPLED_NER_TC_NER_REPAIR_COMPLEX
REACTOME_PRE_NOTCH_TRANSCRIPTION_AND_	25	0.48	0.12	Down	0.002	0.0128
TRANSLATION
ST_GAQ_PATHWAY	24	0.5	0.166667	Down	0.002	0.0128
REACTOME_RORA_ACTIVATES_CIRCADIAN_EXPRESSION	24	0.5	0.041667	Down	0.002	0.0128
REACTOME_ENDOSOMAL_SORTING_COMPLEX_	24	0.083333	0.541667	Up	0.002	0.0128
REQUIRED_FOR_TRANSPORT_ESCRT
BIOCARTA_HDAC_PATHWAY	23	0.478261	0.130435	Down	0.002	0.0128
PID_HDAC_CLASSIII_PATHWAY	22	0.454545	0.136364	Down	0.002	0.0128
PID_RXR_VDR_PATHWAY	22	0.409091	0.045455	Down	0.002	0.0128
REACTOME_PREFOLDIN_MEDIATED_TRANSFER_OF_	21	0.047619	0.571429	Up	0.002	0.0128
SUBSTRATE_TO_CCT_TRIC
REACTOME_SIGNALING_BY_FGFR1_MUTANTS	19	0.421053	0.157895	Down	0.002	0.0128
REACTOME_SIGNALING_BY_FGFR1_FUSION_MUTANTS	18	0.444444	0.111111	Down	0.002	0.0128
BIOCARTA_TNER2_PATHWAY	17	0.529412	0.117647	Down	0.002	0.0128
BIOCARTA_RELA_PATHWAY	15	0.533333	0.2	Down	0.002	0.0128
REACTOME_FORMATION_OF_ATP_BY_CHEMIOSMOTIC_	15	0	0.866667	Up	0.002	0.0128
COUPLING
REACTOME_EARLY_PHASE_OF_HIV_LIFE_CYCLE	13	0	0.538462	Up	0.002	0.0128
BIOCARTA_VDR_PATHWAY	12	0.583333	0	Down	0.002	0.0128
BIOCARTA_CARM1_PATHWAY	12	0.416667	0.166667	Down	0.002	0.0128
REACTOME_SEMA3A_PLEXIN_REPULSION_	12	0.5	0.166667	Down	0.002	0.0128
SIGNALING_BY_INHIBITING_INTEGRIN_ADHESION
BIOCARTA_ETC_PATHWAY	11	0	0.727273	Up	0.002	0.0128
BIOCARTA_EGER_SMRTE_PATHWAY	11	0.454545	0	Down	0.002	0.0128
BIOCARTA_P27_PATHWAY	11	0.090909	0.454545	Up	0.002	0.0128
PID_LPA4_PATHWAY	11	0.545455	0	Down	0.002	0.0128
REACTOME_PURINE_SALVAGE	11	0.181818	0.727273	Up	0.002	0.0128
BIOCARTA_RAB_PATHWAY	10	0	0.9	Up	0.002	0.0128
REACTOME_ASSOCIATION_OF_LICENSING_FACTORS_	9	0.111111	0.555556	Up	0.002	0.0128
WITH_THE_PRE_REPLICATIVE_COMPLEX
REACTOME_GLUTAMATE_NEUROTRANSMITTER_	9	0.555556	0	Down	0.002	0.0128
RELEASE_CYCLE
REACTOME_INTEGRATION_OF_PROVIRUS	8	0	0.625	Up	0.002	0.0128
BIOCARTA_NUCLEARRS_PATHWAY	6	0.5	0	Down	0.002	0.0128
REACTOME_ACYL_CHAIN_REMODELLING_OF_PI	6	0	0.666667	Up	0.002	0.0128
REACTOME_ENDOGENOUS_STEROLS	6	0.5	0.166667	Down	0.002	0.0128
REACTOME_SYNTHESIS_SECRETION_AND_	6	0	0.833333	Up	0.002	0.0128
DEACYLATION_OF_GHRELIN
REACTOME_INTERACTION_BETWEEN_L1_AND_ANKYRINS	6	1	0	Down	0.002	0.0128
KEGG_TAURINE_AND_HYPOTAURINE_METABOLISM	5	0.4	0.2	Down	0.002	0.0128
REACTOME_DOPAMINE_NEUROTRANSMITTER_	5	0.6	0.2	Down	0.002	0.0128
RELEASE_CYCLE
REACTOME_ACETYLCHOLINE_NEUROTRANSMITTER_	4	0.75	0	Down	0.002	0.0128
RELEASE_CYCLE
REACTOME_NUCLEAR_RECEPTOR_TRANSCRIPTION_	34	0.294118	0.058824	Down	0.002	0.0128
PATHWAY
KEGG_PROTEIN_EXPORT	23	0.043478	0.652174	Up	0.002	0.0128
ST_INTERLEUKIN_4_PATHWAY	23	0.391304	0.086957	Down	0.002	0.0128
REACTOME_TRAF6_MEDIATED_IRF7_ACTIVATION	17	0.529412	0	Down	0.002	0.0128
PID_CIRCADIANPATHWAY	15	0.533333	0.066667	Down	0.002	0.0128
REACTOME_VIRAL_MESSENGER_RNA_SYNTHESIS	14	0.071429	0.642857	Up	0.002	0.0128
REACTOME_METABOLISM_OF_POLYAMINES	13	0.076923	0.538462	Up	0.002	0.0128
REACTOME_NOTCH_HLH_TRANSCRIPTION_PATHWAY	11	0.454545	0.090909	Down	0.002	0.0128
REACTOME_ADENYLATE_CYCLASE_ACTIVATING_	7	0.571429	0	Down	0.002	0.0128
PATHWAY
ST_STAT3_PATHWAY	9	0.555556	0	Down	0.002	0.0128
REACTOME_BINDING_AND_ENTRY_OF_HIV_VIRION	4	0	0.5	Up	0.002	0.0128
PID_CD40_PATHWAY	27	0.333333	0.037037	Down	0.002	0.0128
REACTOME_CD28_DEPENDENT_PI3K_AKT_SIGNALING	19	0.473684	0.052632	Down	0.002	0.0128
BIOCARTA_RARRXR_PATHWAY	15	0.4	0.066667	Down	0.002	0.0128
BIOCARTA_PITX2_PATHWAY	13	0.384615	0	Down	0.002	0.0128
REACTOME_INCRETIN_SYNTHESIS_SECRETION_AND_	9	0	0.444444	Up	0.002	0.0128
INACTIVATION
REACTOME_CLASS_C_3_METABOTROPIC_GLUTAMATE_	2	0.5	0	Down	0.002	0.0128
PHEROMONE_RECEPTORS
BIOCARTA_EGF_PATHWAY	31	0.258065	0.032258	Down	0.002	0.0128
REACTOME_HDL_MEDIATED_LIPID_TRANSPORT	11	0.454545	0	Down	0.002	0.0128
REACTOME_GENERIC_TRANSCRIPTION_PATHWAY	292	0.349315	0.10274	Down	0.004	0.0283
REACTOME_DEVELOPMENTAL_BIOLOGY	270	0.333333	0.188889	Down	0.004	0.0283
REACTOME_SIGNALING_BY_PDGF	94	0.361702	0.148936	Down	0.004	0.0283
PID_SMAD2_3NUCLEARPATHWAY	68	0.411765	0.102941	Down	0.004	0.0283
PID_REG_GR_PATHWAY	60	0.366667	0.15	Down	0.004	0.0283
KEGG_ECM_RECEPTOR_INTERACTION	51	0.352941	0.117647	Down	0.004	0.0283
REACTOME_CIRCADIAN_CLOCK	48	0.416667	0.125	Down	0.004	0.0283
KEGG_PPAR_SIGNALING_PATHWAY	43	0.348837	0.162791	Down	0.004	0.0283
SIG_BCR_SIGNALING_PATHWAY	41	0.317073	0.04878	Down	0.004	0.0283
REACTOME_TRANSCRIPTION_COUPLED_NER_TC_NER	41	0.097561	0.439024	Up	0.004	0.0283
REACTOME_RNA_POL_II_TRANSCRIPTION_PRE_	38	0.131579	0.447368	Up	0.004	0.0283
INITIATION_AND_PROMOTER_OPENING
KEGG_AMYOTROPHIC_LATERAL_SCLEROSIS_ALS	37	0.189189	0.324324	Up	0.004	0.0283
KEGG_ABC_TRANSPORTERS	31	0.516129	0.129032	Down	0.004	0.0283
BIOCARTA_PAR1_PATHWAY	31	0.290323	0.16129	Down	0.004	0.0283
REACTOME_COLLAGEN_FORMATION	31	0.451613	0.096774	Down	0.004	0.0283
PID RETINOIC_ACID_PATHWAY	28	0.392857	0.178571	Down	0.004	0.0283
REACTOME_CIRCADIAN_REPRESSION_OF_EXPRESSION_	22	0.5	0.045455	Down	0.004	0.0283
BY_REV_ERBA
KEGG_O_GLYCAN_BIOSYNTHESIS	21	0.047619	0.619048	Up	0.004	0.0283
REACTOME_YAP1_AND_WWTR1_TAZ_STIMULATED_	20	0.4	0.1	Down	0.004	0.0283
GENE_EXPRESSION
BIOCARTA_AKT_PATHWAY	18	0.444444	0.166667	Down	0.004	0.0283
BIOCARTA_IL7_PATHWAY	16	0.4375	0.125	Down	0.004	0.0283
REACTOME_OXYGEN_DEPENDENT_PROLINE_	15	0.066667	0.533333	Up	0.004	0.0283
HYDROXYLATION_OF_HYPOXIA_INDUCIBLE_FA
BIOCARTA_IL22BP_PATHWAY	14	0.5	0	Down	0.004	0.0283
REACTOME_NCAM1_INTERACTIONS	14	0.571429	0	Down	0.004	0.0283
REACTOME_EFFECTS_OF_PIP2_HYDROLYSIS	14	0.428571	0.071429	Down	0.004	0.0283
KEGG_RIBOFLAVIN_METABOLISM	13	0.076923	0.461538	Up	0.004	0.0283
REACTOME_TRAF3_DEPENDENT_IRF_ACTIVATION_	13	0.461538	0	Down	0.004	0.0283
PATHWAY
BIOCARTA_EPONFKB_PATHWAY	9	0.666667	0	Down	0.004	0.0283
REACTOME_IL_6_SIGNALING	9	0.444444	0	Down	0.004	0.0283
REACTOME_SYNTHESIS_SECRETION_AND_	7	0	0.571429	Up	0.004	0.0283
INACTIVATION_OF_GIP
BIOCARTA_GABA_PATHWAY	3	0	0.666667	Up	0.004	0.0283
REACTOME_INFLUENZA_VIRAL_RNA_TRANSCRIPTION_	98	0.020408	0.867347	Up	0.004	0.0283
AND_REPLICATION
REACTOME_LIPOPROTEIN_METABOLISM	19	0.315789	0.052632	Down	0.004	0.0283
REACTOME_ACYL_CHAIN_REMODELLING_OF_PG	7	0	0.571429	Up	0.004	0.0283
BIOCARTA_PDGF_PATHWAY	30	0.266667	0.033333	Down	0.004	0.0283
REACTOME_SYNTHESIS_SECRETION_AND_	8	0	0.5	Up	0.004	0.0283
INACTIVATION_OF_GLP1
BIOCARTA_SALMONELLA_PATHWAY	11	0	0.636364	Up	0.004	0.0283
REACTOME_AXON_GUIDANCE	173	0.34104	0.179191	Down	0.006	0.0386
REACTOME_SIGNALING_BY_NOTCH	90	0.311111	0.2	Down	0.006	0.0386
KEGG_PEROXISOME	71	0.098592	0.352113	Up	0.006	0.0386
ST_INTEGRIN_SIGNALING_PATHWAY	71	0.323944	0.126761	Down	0.006	0.0386
REACTOME_SEMAPHORIN_INTERACTIONS	58	0.310345	0.224138	Down	0.006	0.0386
REACTOME_RNA_POL_II_PRE_TRANSCRIPTION_EVENTS	57	0.175439	0.385965	Up	0.006	0.0386
KEGG_ACUTE_MYELOID_LEUKEMIA	53	0.320755	0.132075	Down	0.006	0.0386
REACTOME_NUCLEOTIDE_EXCISION_REPAIR	46	0.108696	0.391304	Up	0.006	0.0386
REACTOME_EXTRACELLULAR_MATRIX_ORGANIZATION	43	0.372093	0.093023	Down	0.006	0.0386
KEGG_VALINE_LEUCINE_AND_ISOLEUCINE_	40	0.075	0.45	Up	0.006	0.0386
DEGRADATION
PID_HDAC_CLASII_PATHWAY	31	0.419355	0.16129	Down	0.006	0.0386
REACTOME_ELONGATION_ARREST_AND_RECOVERY	31	0.193548	0.451613	Up	0.006	0.0386
KEGG_RNA_POLYMERASE	27	0.074074	0.481481	Up	0.006	0.0386
SIG_IL4RECEPTOR_IN_B_LYPHOCYTES	25	0.32	0.04	Down	0.006	0.0386
PID_REELINPATHWAY	24	0.416667	0.166667	Down	0.006	0.0386
REACTOME_ABC_FAMILY_PROTEINS_MEDIATED_	23	0.521739	0.217391	Down	0.006	0.0386
TRANSPORT
REACTOME_ABORTIVE_ELONGATION_OF_HIV1_	23	0.130435	0.478261	Up	0.006	0.0386
TRANSCRIPT_IN_THE_ABSENCE_OF_TAT
BIOCARTA_GH_PATHWAY	22	0.363636	0.045455	Down	0.006	0.0386
REACTOME_RNA_POL_III_CHAIN_ELONGATION	16	0.0625	0.4375	Up	0.006	0.0386
BIOCARTA_CD40_PATHWAY	14	0.5	0.071429	Down	0.006	0.0386
REACTOME_ACYL_CHAIN_REMODELLING_OF_PC	12	0.166667	0.5	Up	0.006	0.0386
REACTOME_CASPASE_MEDIATED_CLEAVAGE_OF_	11	0.545455	0.272727	Down	0.006	0.0386
CYTOSKELETAL_PROTEINS
REACTOME_ORGANIC_CATION_ANION_ZWITTERION_	5	0.6	0	Down	0.006	0.0386
TRANSPORT
KEGG_FOCAL_ADHESION	145	0.296552	0.151724	Down	0.006	0.0386
PID_TNFPATHWAY	43	0.395349	0.093023	Down	0.006	0.0386
REACTOME_APC_CDC20_MEDIATED_DEGRADATION_OF_	18	0.111111	0.388889	Up	0.006	0.0386
NEK2A
BIOCARTA_ETS_PATHWAY	17	0.352941	0.117647	Down	0.006	0.0386
PID_HIF1APATHWAY	18	0.166667	0.333333	Up	0.006	0.0386
KEGG_TRYPTOPHAN_METABOLISM	25	0.08	0.28	Up	0.006	0.0386
REACTOME_N_GLYCAN_ANTENNAE_ELONGATION	10	0.1	0.5	Up	0.006	0.0386
REACTOME_AMINO_ACID_TRANSPORT_ACROSS_THE_	18	0.388889	0	Down	0.006	0.0386
PLASMA_MEMBRANE

indicates data missing or illegible when filed

Dataset 3. GSEA results detailing lung cancer associated dataset enrichment among genes differentially expressed in

the airway field associated with PMLs

				NOM	FDR	FWER	RANK
Gene Set	SIZE	ES	NES	p-val	q-val	p-val	AT MAX	LEADING EDGE

OOI ET AL. EARLY, DN-REG,	26	−0.56	−1.87	0.002	0.005	0.017	2634	tags = 46%, list = 19%, signal = 57%
PVN P < 0.05, TVN P < 0.05
OOI ET AL. EARLY, UP-REG,	487	0.36	2.11	0	0	0.001	3850	tags = 43%, list = 28%, signal = 58%
PVN P < 0.05, TVN P < 0.05
OOI ET AL. STEPWISE, DN-REG,	111	−0.31	−1.4	0.028	0.064	0.794	3041	tags = 27%, list = 22%, signal = 54%
PVN P < 0.05, TVP P < 0.05,
TVN P < 0.05
OOI ET AL. STEPWISE, UP-REG,	518	0.29	1.73	0.	0.005	0.076	2858	tags = 29%, list = 21%, signal = 35%
PVN P < 0.05, TVP P < 0.05,
TVN P < 0.05
OOI ET AL. LATE, DN-REG,	12	−0.64	−1.74	0.012	0.009	0.082	1784	tags = 58%, list = 13%, signal = 67%
TVP P < 0.05, TVN P < 0.05
OOI ET AL. LATE, UP-REG,	54	0.53	2.24	0	0	0	3052	tags = 46%, list = 22%, signal = 59
TVP P < 0.05, TVN P < 0.05
TCGA, SCCVN, DN-REG, 200	119	−0.37	−1.67	0.001	0.014	0.152	3526	tags = 36%, list = 25%, signal = 48%
TCGA, SCCVN, UP-REG, 200	146	0.28	1.41	0.013	0.048	0.6	3950	tags = 40%, list = 28%, signal = 55%
GSE18842, TVN, DN-REG, 200	111	−0.42	−1.87	0	0.007	0.016	3526	tags = 41%, list = 25%, signal = 54%
GSE18842, TVN, UP-REG, 200	149	0.43	2.14	0	0	0.001	4601	tags = 52%, list = 33%, signal = 77%
GSE19188, SCCVN, DN-REG, 200	115	−0.35	−1.55	0.006	0.027	0.371	4837	tags = 50%, list = 35%, signal = 75%
GSE19188, SCCVN, UP-REG, 200	147	0.42	2.14	0	0	0.001	3596	tags = 41%, list = 26%, signal = 55%
GSE4115, CAVN, DN-REG, 200	108	−0.35	−1.56	0.005	0.031	0.365	3066	tags = 31%, list = 22%, signal = 39%
GSE4115, CAVN, UP-REG, 200	197	0.45	2.36	0	0	0	3781	tags = 55%, list = 27%, signal = 74%

REFERENCES

1. Hackett N R, Heguy A, Harvey B G, et al. (2003) Variability of antioxidant-related gene expression in the airway epithelium of cigarette smokers. American journal of respiratory cell and molecular biology 29(3 Pt 1):331-343.
2. Harvey B G, Heguy A, Leopold P L, et al. (2007) Modification of gene expression of the small airway epithelium in response to cigarette smoking. J Mol Med (Berl) 85(1):39-53.
3. Beane J, Sebastiani P, Liu G, et al. (2007) Reversible and permanent effects of tobacco smoke exposure on airway epithelial gene expression. Genome biology 8(9):1-17.
4. Spira A, Beane J, Shah V, et al. (2004) Effects of cigarette smoke on the human airway epithelial cell transcriptome. Proceedings of the National Academy of Sciences of the United States of America 101(27):10143-10148.
5. Beane J, Vick J, Schembri F, et al. (2011) Characterizing the impact of smoking and lung cancer on the airway transcriptome using RNA-Seq. Cancer Prev Res (Phila) 4(6):803-817.
6. Sridhar S, Schembri F, Zeskind J, et al. (2008) Smoking-induced gene expression changes in the bronchial airway are reflected in nasal and buccal epithelium. BMC genomics 9:259.
7. Schembri F, Sridhar S, Perdomo C, et al. (2009) MicroRNAs as modulators of smoking-induced gene expression changes in human airway epithelium. Proceedings of the National Academy of Sciences of the United States of America 106(7):2319-2324.
8. Chari R, Lonergan K M, Ng R T, et al. (2007) Effect of active smoking on the human bronchial epithelium transcriptome. BMC genomics 8:297.
9. Spira A, Beane J E, Shah V, et al. (2007) Airway epithelial gene expression in the diagnostic evaluation of smokers with suspect lung cancer. Nature medicine 13(3):361-366.
10. Beane J, Sebastiani P, Whitfield T H, et al. (2008) A prediction model for lung cancer diagnosis that integrates genomic and clinical features. Cancer Prev Res (Phila) 1(1):56-64.
11. Whitney D H, Elashoff M R, Porta-Smith K, et al. (2015) Derivation of a bronchial genomic classifier for lung cancer in a prospective study of patients undergoing diagnostic bronchoscopy. BMC medical genomics 8:18.
12. Silvestri G A, Vachani A, Whitney D, et al. (2015) A Bronchial Genomic Classifier for the Diagnostic Evaluation of Lung Cancer. The New England journal of medicine 373(3):243-251.
13. Gustafson A M, Soldi R, Anderlind C, et al. (2010) Airway PI3K pathway activation is an early and reversible event in lung cancer development. Science translational medicine 2(26):26ra25.
14. Wistuba, I I & Gazdar A F (2006) Lung cancer preneoplasia. Annual review of pathology 1:331-348.
15. Wistuba, I I, Lam S, Behrens C, et al. (1997) Molecular damage in the bronchial epithelium of current and former smokers. Journal of the National Cancer Institute 89(18):1366-1373.
16. Wistuba, I I, Behrens C, Virmani A K, et al. (2000) High resolution chromosome 3p allelotyping of human lung cancer and preneoplastic/preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. Cancer research 60(7):1949-1960.
17. Wistuba, I I, Behrens C, Milchgrub S, et al. (1999) Sequential molecular abnormalities are involved in the multistage development of squamous cell lung carcinoma. Oncogene 18(3):643-650.
18. Belinsky S A, Palmisano W A, Gilliland F D, et al. (2002) Aberrant promoter methylation in bronchial epithelium and sputum from current and former smokers. Cancer research 62(8):2370-2377.
19. Lamy A, Sesboue R, Bourguignon J, et al. (2002) Aberrant methylation of the CDKN2a/p16INK4a gene promoter region in preinvasive bronchial lesions: a prospective study in high-risk patients without invasive cancer. International journal of cancer 100(2):189-193.
20. Nakachi I, Rice J L, Coldren C D, et al. (2014) Application of SNP microarrays to the genome-wide analysis of chromosomal instability in premalignant airway lesions. Cancer Prev Res (Phila) 7(2):255-265.
21. Rahman S M, Gonzalez A L, Li M, et al. (2011) Lung cancer diagnosis from proteomic analysis of preinvasive lesions. Cancer research 71(8):3009-3017.
22. Massion P P, Zou Y, Uner H, et al. (2009) Recurrent genomic gains in preinvasive lesions as a biomarker of risk for lung cancer. PloS one 4(6):e5611.
23. van Boerdonk R A, Sutedja T G, Snijders P J, et al. (2011) DNA copy number alterations in endobronchial squamous metaplastic lesions predict lung cancer. American journal of respiratory and critical care medicine 184(8):948-956.
24. Ishizumi T, McWilliams A, MacAulay C, Gazdar A, & Lam S (2010) Natural history of bronchial preinvasive lesions. Cancer metastasis reviews 29(1):5-14.
25. Lam S, Kennedy T, Unger M, et al. (1998) Localization of bronchial intraepithelial neoplastic lesions by fluorescence bronchoscopy. Chest 113(3):696-702.
26. Edell E, Lam S, Pass H, et al. (2009) Detection and localization of intraepithelial neoplasia and invasive carcinoma using fluorescence-reflectance bronchoscopy: an international, multicenter clinical trial. Journal of thoracic oncology: official publication of the International Association for the Study of Lung Cancer 4(1):49-54.
27. van Boerdonk R A, Smesseim I, Heideman D A, et al. (2015) Close Surveillance with Long-Term Follow-up of Subjects with Preinvasive Endobronchial Lesions. American journal of respiratory and critical care medicine 192(12):1483-1489.
28. Jeremy George P, Banerjee A K, Read C A, et al. (2007) Surveillance for the detection of early lung cancer in patients with bronchial dysplasia. Thorax 62(1):43-50.
29. Tammemagi M C, Lam S C, McWilliams A M, & Sin D D (2011) Incremental value of pulmonary function and sputum DNA image cytometry in lung cancer risk prediction. Cancer Prev Res (Phila) 4(4):552-561.
30. Li B & Dewey C N (2011) RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC bioinformatics 12:323.
31. Langmead B, Trapnell C, Pop M, & Salzberg S L (2009) Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10(3):R25.
32. Katz Y, Wang E T, Airoldi E M, & Burge C B (2010) Analysis and design of RNA sequencing experiments for identifying isoform regulation. Nature methods 7(12):1009-1015.
33. Anders S, Pyl P T, & Huber W (2014) HTSeq—A Python framework to work with high-throughput sequencing data. bioRxiv.
34. Piccolo S R, Sun Y, Campbell J D, et al. (2012) A single-sample microarray normalization method to facilitate personalized-medicine workflows. Genomics 100(6):337-344.
35. Ritchie M E, Phipson B, Wu D, et al. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 43(7):gkv007-e047.
36. Robinson M D, McCarthy D J, & Smyth G K (2010) edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Journal of Gerontology 26(1):139-140.
37. Leek J T, Johnson W E, Parker H S, Jaffe A E, & Storey J D (2012) The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics (Oxford, England) 28(6):882-883.
38. Law C W, Chen Y, Shi W, & Smyth G K (2014) Voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome biology.
39. Wu D, Lim E, Vaillant F, et al. (2010) ROAST: rotation gene set tests for complex microarray experiments. Bioinformatics 26(17):2176-2182.
40. Subramanian A, Tamayo P, Mootha V K, et al. (2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America 102(43):15545-15550.
41. Hanzelmann S, Castelo R, & Guinney J (2013) GSVA: gene set variation analysis for microarray and RNA-seq data. BMC bioinformatics 14:7.
42. Mazzilli S A, Hershberger P A, Reid M E, et al. (2015) Vitamin D Repletion Reduces the Progression of Premalignant Squamous Lesions in the NTCU Lung Squamous Cell Carcinoma Mouse Model. Cancer Prev Res (Phila) 8(10):895-904.
43. Chacko B K, Kramer P A, Ravi S, et al. (2014) The Bioenergetic Health Index: a new concept in mitochondrial translational research. Clin Sci (Lond) 127(6):367-373.
44. Ooi A T, Gower A C, Zhang K X, et al. (2014) Molecular profiling of premalignant lesions in lung squamous cell carcinomas identifies mechanisms involved in stepwise carcinogenesis. Cancer Prev Res (Phila) 7(5):487-495.
45. Dang C V (2012) Links between metabolism and cancer. Genes & development 26(9):877-890.
46. Chen X, Qian Y, & Wu S (2015) The Warburg effect: evolving interpretations of an established concept. Free radical biology & medicine 79:253-263.
47. Phelan J J, MacCarthy F, Feighery R, et al. (2014) Differential expression of mitochondrial energy metabolism profiles across the metaplasia-dysplasia-adenocarcinoma disease sequence in Barrett's oesophagus. Cancer letters 354(1):122-131.
48. Xylas J, Varone A, Quinn K P, et al. (2015) Noninvasive assessment of mitochondrial organization in three-dimensional tissues reveals changes associated with cancer development. International journal of cancer 136(2):322-332.
49. Grimm M, Cetindis M, Lehmann M, et al. (2014) Association of cancer metabolism-related proteins with oral carcinogenesis—indications for chemoprevention and metabolic sensitizing of oral squamous cell carcinoma? Journal of translational medicine 12:208.
50. Pan J, Zhang Q, Liu Q, et al. (2014) Honokiol inhibits lung tumorigenesis through inhibition of mitochondrial function. Cancer Prev Res (Phila) 7(11):1149-1159.
51. Zhang X, Sebastiani P, Liu G, et al. (2010) Similarities and differences between smoking-related gene expression in nasal and bronchial epithelium. Physiological genomics 41(1):1-8.
52. Campbell J D, Mazzilli S A, Reid M E, et al. (2016) The Case for a Pre-Cancer Genome Atlas (PCGA). Cancer Prev Res (Phila) 9(2):119-124.
53. Kensler T W, Spira A, Garber J E, et al. (2016) Transforming Cancer Prevention through Precision Medicine and Immune-oncology. Cancer Prev Res (Phila) 9(1):2-10.
54. Edgar R, Domrachev M, & Lash A E (2002) Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Research 30(1):207-210.
55. Irizarry R A, Hobbs B, Collin F, et al. (2003) Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics (Oxford, England) 4(2):249-264.
56. Golub T R, Slonim D K, Tamayo P, et al. (1999) Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286(5439):531-537.
57. Johnson W E, Li C, & Rabinovic A (2006) Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics (Oxford, England) 8(1):118-127.
58. Wang L, Wang S, & Li W (2012) RSeQC: quality control of RNA-seq experiments. Bioinformatics (Oxford, England) 28(16):2184-2185.
59. Trapnell C, Williams B A, Pertea G, et al. (2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nature biotechnology 28(5):511-515.
60. Robinson M D & Oshlack A (2010) A scaling normalization method for differential expression analysis of RNA-seq data. Genome biology 11(3):R25.
61. Friedman J, Hastie T, & Tibshirani R (2010) Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of statistical software 33(1):1-968.
62. Liaw A & Wiener M (2002) Classification and regression by randomForest. R news.
63. Anders S & Huber W (2010) Differential expression analysis for sequence count data. Genome biology 11(10):1.
64. Buja A & Eyuboglu N (1992) Remarks on Parallel Analysis. Multivariate behavioral research 27(4):509-540.
65. McClish D K (1989) Analyzing a portion of the ROC curve. Medical decision making: an international journal of the Society for Medical Decision Making 9(3):190-195.
66. Gentleman R, Carey V, Huber W, & Hahne F (2015) Genefilter: methods for filtering genes from high-throughput experiments (R package version).
67. Meyer D, Dimitriadou E, Hornik K, Weingessel A, & Leisch F (2015) e1071: Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), TU Wien package e1071 version 1.6-71. (Comprehensive R Archive Network (CRAN)).
68. Robin X, Turck N, Hainard A, et al. (2011) pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC bioinformatics 12(1):77.

Claims

1.-18. (canceled)

19. A method of processing a sample from a subject suspected of having a premalignant bronchial lesion comprising the steps of: (a) providing a biological sample from the mouth or nose of the subject or from a brushing of the bronchi walls of the subject; and (b) measuring the expression of two or more genes in the sample by northern-blot hybridization, a ribonuclease protection assay, or a reverse transcriptase polymerase chain reaction (RT-PCR) method, wherein the two or more genes are genes involved in an oxidative phosphorylation (OXPHOS), electron transport chain (ETC), or mitochondrial protein transport pathway.

20. The method of claim 19, wherein the expression of at least five genes involved in an oxidative phosphorylation (OXPHOS), electron transport chain (ETC), or mitochondrial protein transport pathway are measured.

21. The method of claim 19, wherein the expression in the sample of at least twenty genes are measured.

22. The method of claim 19, wherein the two or more genes comprise cDNA.

23. The method of claim 19, wherein the expression of two or more genes in the sample is measured by an RT-PCR method.

24. The method of claim 19, wherein the biological sample is obtained from the mouth of the subject.

25. The method of claim 19, wherein the subject has a positive result in an imaging study of a premalignant bronchial lesion.

26. The method of claim 19, wherein the subject has previously been diagnosed with a lung, bronchus, head/neck, and/or esophagus cancer but has no current evidence of the cancer.

27. The method of claim 19, wherein the subject is a current smoker or a former smoker with 20+ pack years.

28. The method of claim 27, wherein the subject is at least 50 years old.

29. The method of claim 19, wherein the subject has emphysema, chronic bronchitis, chronic obstructive pulmonary disease, an occupationally related asbestos disease, or a family history of lung cancer in a first degree relative.

30. The method of claim 29, wherein the subject is at least 50 years old.