UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Circulating microRNAs as biomarkers for early cancer detection MacLellan, Sara Ann 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_maclellan_sara.pdf [ 1.74MB ]
Metadata
JSON: 24-1.0073891.json
JSON-LD: 24-1.0073891-ld.json
RDF/XML (Pretty): 24-1.0073891-rdf.xml
RDF/JSON: 24-1.0073891-rdf.json
Turtle: 24-1.0073891-turtle.txt
N-Triples: 24-1.0073891-rdf-ntriples.txt
Original Record: 24-1.0073891-source.json
Full Text
24-1.0073891-fulltext.txt
Citation
24-1.0073891.ris

Full Text

CIRCULATING MICRORNAS AS BIOMARKERS FOR EARLY CANCER DETECTION  by  Sara Ann MacLellan BSc., St. Francis Xavier University, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  The Faculty of Graduate Studies (Interdisciplinary Oncology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2013 © Sara Ann MacLellan, 2013  Abstract Cancer is a major cause of death worldwide. Patients diagnosed at an early stage have an improved prognosis and therefore efforts have been made into the development of methods to detect tumors at their earliest stages. MicroRNAs (miRNAs) are non-coding RNAs that negatively regulate gene expression by interfering with the translation of target mRNAs. Studies have found that miRNAs are present at stable levels in the circulation and that they are differentially expressed in patients with various diseases. In this thesis we used qRT-PCR to assess the utility of 742 serum miRNAs as biomarkers for early cancer detection. In aim 1 we examined the levels of serum miRNAs in patients with high-risk oral lesions. We identified five miRNAs that are significantly deregulated in the serum of these patients compared to demographically matched, non-cancer controls. Additionally, these miRNAs correspondingly decreased or increased after surgical resection of the lesion. In aim 2 we examined the effect of hemolysis, fasting, and smoking on the serum miRNA levels of healthy individuals. We also compared serum miRNA profiles of samples taken from healthy individuals over different time periods. We found that mechanical hemolysis of blood samples simulating blood drawing can significantly alter serum miRNA quantification and should be taken into consideration when identifying endogenous controls and candidate biomarkers for circulating miRNA studies. Fasting, smoking, and a time period up to 17 months between samples were demonstrated to not have a significant effect on the overall serum miRNA profiles of healthy individuals. In aim 3 we compared the miRNA profiles of paired samples collected during surgery from the same patient from a) pulmonary venous effluent draining the tumor vascular bed (tumor associated samples) and b) systemic arterial blood to identify lung adenocarcinoma biomarkers. We found 35 ii  miRNAs that were significantly up-regulated in tumor-associated serum samples. However, when we tested the candidate miRNAs in cancer versus non-cancer peripheral venous blood samples they were not significantly differentially expressed. The results presented in this thesis demonstrate the need for standardized protocols for circulating miRNA studies and provide evidence for the utility of serum miRNAs as biomarkers of disease.  iii  Preface  Portions of this thesis have been published as described below.  A version of chapter 3 has been published. MacLellan SA, Lawson J, Baik J, Guillaud M, Poh CF , and Garnis C. (2012) Differential expression of miRNAs in the serum of patients with highrisk oral lesions. Cancer Medicine 1(2): 268-274. I conducted the majority of the RNA extractions and qRT-PCR profiling with the assistance of J Lawson. I also wrote the manuscript and conducted the data analysis with the exception of the ROC curves which were made by J Baik and M Guillaud. CF Poh was the oral pathologist for the study and contributed to study design. C Garnis conceived and designed the study.  A version of chapter 4 has been submitted for publication. MacLellan SA, MacAulay C, Lam S, and Garnis C. (2013) Pre-profiling factors influencing serum microRNA levels. I conducted the majority of the blood processing and all of the RNA extractions and qRT-PCR profiling for the study. I also contributed to study design and wrote the manuscript. I conducted the data analysis with the help of C MacAulay. S Lam, C MacAulay, and C Garnis contributed to the study conception and design.  In the above chapters, edits have been made to the published version to better integrate the work into the thesis and clarify material where necessary. Most changes are minor with the exception of the introduction of each manuscript which has been incorporated into chapter 1 of this thesis.  iv  Chapter 5 contains work from our collaborators. Dr. Stephen Lam and Dr. John Yee contributed to the study conception and design and provided all serum samples used in the study. Dr. Cathie Garnis performed RNA extractions for the surgical samples and contributed to study conception and design. I conducted all other RNA extractions and all qRT-PCR profiling with the exception of the surgical samples that were profiled by Exiqon qRT-PCR Services. I conducted all data analyses.  The work described in this thesis was approved by the Research Ethics Board of The University of British Columbia. Ethics Certificate Number: H10-02846.  v  Table of contents Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iv Table of contents ............................................................................................................................ vi List of Tables ................................................................................................................................. ix List of Figures ................................................................................................................................. x List of Abbreviations ..................................................................................................................... xi Acknowledgements ....................................................................................................................... xii Dedication .................................................................................................................................... xiii 1. Introduction ................................................................................................................................. 1 1.1 Biomarkers for the early detection of cancer ........................................................................ 1 1.2 Early detection of oral cancer................................................................................................ 2 1.3 Early detection of lung cancer ............................................................................................... 4 1.4 microRNAs............................................................................................................................ 5 1.5 miRNAs and cancer .............................................................................................................. 8 1.6 Circulating miRNAs ............................................................................................................ 10 1.7 Circulating miRNAs and oral cancer .................................................................................. 13 1.8 Circulating miRNAs and lung cancer ................................................................................. 14 1.9 Challenges associated with circulating miRNA biomarker studies .................................... 17 1.10 Rationale, hypotheses and aims ........................................................................................ 20 2. Materials and methods .............................................................................................................. 22 2.1 Sample collection ................................................................................................................ 22 2.2 Blood processing ................................................................................................................. 25 2.3 Hemoglobin concentration .................................................................................................. 26 vi  2.4 RNA purification ................................................................................................................. 26 2.5 miRNA quantification by qRT-PCR ................................................................................... 27 2.6 Data analysis ....................................................................................................................... 27 3. Differential expression of miRNAs in the serum of patients with high-risk oral lesions ......... 28 3.1 Introduction ......................................................................................................................... 28 3.2 Results ................................................................................................................................. 29 3.2.1 miRNA expression in patients with OSCC and CIS .................................................... 29 3.2.2 Serum miRNA levels pre and post surgical resection of an oral HRL ......................... 33 3.2.3. Serum miRNA biomarker candidates .......................................................................... 35 3.2.4 Correlation between oral HRL tissue and serum miRNA profiles ............................... 38 3.3 Discussion ........................................................................................................................... 38 4. Pre-profiling factors influencing serum microRNA levels.......................................................... 42 4.1 Introduction ......................................................................................................................... 42 4.2 Results ................................................................................................................................. 43 4.2.1 The effects of hemolysis on serum miRNA levels ....................................................... 43 4.2.2 Endogenous controls for serum miRNA studies .......................................................... 49 4.2.3 The effects of fasting on serum miRNA profiles of healthy individuals ...................... 50 4.2.4 The effects of smoking on serum miRNA profiles....................................................... 52 4.2.5 Changes in serum miRNA levels over time ................................................................. 54 4.3 Discussion ........................................................................................................................... 55 5. Identification of differentially expressed miRNAs in the serum of patients with early lung adenocarcinoma ............................................................................................................................ 61 5.1 Introduction ......................................................................................................................... 61 5.2 Results ................................................................................................................................. 61 5.2.1 Identification of differentially expressed miRNAs in patient-matched samples .......... 61 vii  5.2.2 Comparison of miRNA profiles from patients with lung adenocarcinoma and noncancer controls ....................................................................................................................... 65 5.2.3 Overlap between surgical and peripheral venous sample sets ...................................... 68 5.3 Discussion ........................................................................................................................... 71 6. Review of aims ......................................................................................................................... 73 7. Conclusion ................................................................................................................................ 75 7.1 Future perspectives .............................................................................................................. 75 7.2 Conclusion........................................................................................................................... 79 References ..................................................................................................................................... 80  viii  List of Tables Table 1: Current studies on the use of serum miRNAs as biomarkers of NSCLC .......................15 Table 2: Aim 1 patient demographics ...........................................................................................24 Table 3: Aim 3 patient demographics ...........................................................................................25 Table 4: Results of a one-way ANOVA followed by Tukey’s HSD post hoc on oral HRL samples and controls ......................................................................................................................31 Table 5: miRNAs significantly (P > 0.001) up or down-regulated in oral cancer serum samples as determined by a permutation test...............................................................................................32 Table 6: Hemoglobin concentrations of matched lysed and unlysed serum samples ...................44 Table 7: List of 144 miRNAs most affected by mechanical hemolysis........................................46 Table 8: The correlation of potential endogenous controls to spiked-in synthetic miR-cel-39-3p levels ..............................................................................................................................................50 Table 9: Triglyceride concentration of fasting and non-fasting serum samples ...........................51 Table 10: Patient demographics for smoking/non-smoking samples ..........................................53 Table 11: Pearson correlation coefficients for miRNA profiles of serum collected at two different time points from healthy individuals ..............................................................................55 Table 12: Results of a fold-change analysis of pulmonary venous and systemic arterial lung cancer samples ..............................................................................................................................63 Table 13: miRNAs significantly (uncorrected P-value < 0.05) up or down-regulated in lung adenocarcinoma serum samples compared to controls .................................................................68 Table 14: Average normalized Cts from peripheral venous samples of the miRNAs most significantly up-regulated in surgical samples ..............................................................................71  ix  List of Figures Figure 1: miRNA biogenesis and incorporation into the RISC complex ......................................8 Figure 2: Current theories on how miRNAs enter circulation via exosomes, microparticles, proteins, and lipoproteins ..............................................................................................................13 Figure 3: An average-linkage hierarchical clustering analysis of 50 miRNAs with the highest standard deviation across all pre- vs. post-surgery cases ..............................................................34 Figure 4: (A) Expression levels of the five candidate miRNAs in the serum of oral HRL and non-cancer control patients. (B) ROC curve analysis conducted on the cancer versus control cases showing the five most significantly deregulated miRNAs in the serum of patients with oral HRLs .............................................................................................................................................36 Figure 5: An average-linkage Pearson correlation cluster analysis of the serum miRNA profiles of matched unlysed and lysed samples .........................................................................................48 Figure 6: The total number of miRNAs present at detectable levels in fasting and non-fasting samples ..........................................................................................................................................52 Figure 7: Levels of miR-128 in the serum of non-smokers and current smokers .......................53 Figure 8: The number of miRNAs detected in pulmonary venous and systemic arterial blood samples after removal of miRNAs significantly affected by hemolysis .......................................65 Figure 9: An average-linkage Pearson correlation cluster analysis of the serum miRNA profiles of lung adenocarcinoma patients and non-cancer controls ...........................................................67 Figure 10: (A) miR-532 is up-regulated in pulmonary venous samples compared to matched systemic arterial samples. (B) miR-532-5p is also slightly up-regulated in lung adenocarcinoma peripheral venous samples compared to controls .........................................................................70  x  List of Abbreviations miRNA  microRNA  OSCC  oral squamous cell carcinoma  NSCLC  Non-small cell lung cancer  CT  Computed tomography  Pre-miRNA  pre-miRNA  Exp5  Exportin-5  RISC  RNA-induced silencing complex  Argo  Argonaute  HDL  High density lipoprotein  CIS  Carcinoma in situ  AC  Adenocarcinoma  SCC  Squamous cell carcinoma  LCC  Large cell carcinoma  qRT-PCR  quantitative reverse transcription real-time polymerase chain reaction  HRL  High risk lesion  ROC  Receiver operating characteristic  AUC  Area under the curve  dPCR  Digital polymerase chain reaction  COPD  Chronic obstructive pulmonary disease  xi  Acknowledgements Firstly, I would like to thank my graduate supervisor Dr. Cathie Garnis for her guidance and patience throughout my degree. Working in her lab has been a wonderful experience and I’ve learned so many skills, both in the lab and outside of it, which I will keep with me as I continue my career.  I would also like to thank my supervisory committee, Dr. Stephen Lam and Dr. Calum MacAulay for all their help throughout this thesis. It has been a privilege to work with experts in this field.  I also want to thank all of the collaborators, fellow members of the Garnis lab, and volunteers who helped with the work described in this thesis. The studies presented here would not have been possible without their support and contributions.  I would like to acknowledge the financial support of the University of British Columbia Interdisciplinary Oncology Program, the Pacific Otolaryngology Foundation, and the Terry Fox Research Institute.  xii  Dedication To my parents, my brothers, and the rest of my amazing friends and family. Thank you so much for your support and encouragement throughout this whole process.  xiii  1. Introduction  1.1 Biomarkers for the early detection of cancer Cancer is a devastating disease and a major cause of death worldwide accounting for ~ 13% of all deaths in 20081. Patients diagnosed at an early stage have an improved prognosis compared to those diagnosed at a late stage and early detection can reduce cancer mortality2. Thus, a considerable amount of resources have been invested in the development of efficient screening methods to detect tumors at their earliest stages3. To develop screening programs many researchers are conducting studies dedicated to the identification of highly sensitive and specific tumor biomarkers4. The variety of proteins, nucleic acids, and other molecular elements in blood has made it a popular medium for such studies5. The major advantage of blood-based biomarkers is that they can be detected non-invasively5. In addition to early detection, bloodbased biomarkers may also be useful for cancer diagnosis, monitoring progression and response to treatment, and for identifying disease recurrence. Despite an increase in publications of potentially useful non-invasive cancer biomarkers, very few biomarkers are used in standard clinical practice6. This could be due to a variety of problems that can occur during the pre-analytical, analytical, and post-analytical stages of biomarker discovery and validation7. Blood-based cancer biomarkers currently used in the clinic are all proteins including: α-fetoprotein and human chorionic gonadotropin-β for testicular cancer, prostate-specific antigen for prostate cancer, and CA-125 for ovarian cancer8. However, even these biomarkers have limitations. For example, levels of prostate-specific antigen are elevated by a number of pathologies leading to a reduced specificity for cancer detection8.  1  1.2 Early detection of oral cancer Oral cancer is a cancer of the oral cavity or the oropharynx. Oral cancer is one of the most commonly diagnosed cancers worldwide and in high-risk countries (e.g. in South-East Asia) this disease represents ~25% of all new cancers9. Among the different subtypes of oral cancer, oral squamous cell carcinoma (OSCC) is the most common contributing to over 90% of oral malignancies9. OSCC carries a poor prognosis and, despite advances in treatment, overall survival rates for patients have seen little improvement. Early diagnosis is a major factor contributing to the survival rates of OSCC patients as those diagnosed at an early stage have a five-year survival rate of about 80% to 90% but for patients diagnosed at an advanced stage of disease, the five-year survival rate drops to about 30%10. A high incidence of recurrence is also a major factor contributing to the poor survival rates of patients diagnosed with OSCC with about 30% of males and 20% of females developing secondary primary tumors11. Although OSCC is preceded by visible lesions in the oral mucosa and the oral cavity is easily accessed for clinical inspection, ~ 64% of OSCCs are detected at a late stage with ~ 50% exhibiting regional cervical metastases12, 13. Studies examining the reasons for this delay in detection often cite a lack of public awareness as well as a need for more action on the part of health-care providers14. Therefore, in order to improve early detection rates and outcomes for OSCC patients the role of oral cancer screening programs have been examined15. A study in 2005 conducted over 9 years on over 130 000 participants provided evidence on the positive impact of screening high-risk individuals (those with high tobacco and/or alcohol intake). The results of this study suggested that oral cancer screening by visual inspection of the oral cavity has the potential to prevent ~37 000 oral cancer deaths worldwide16.  2  In addition to visual inspection other minimally invasive techniques used for oral cancer screening include: brush biopsy, toluidine blue staining, and tissue fluorescence imaging14. Brush biopsies involve collecting a complete transepithelial specimen from mucosal lesions that appear clinically benign and would not otherwise receive a scalpel biopsy17. Studies examining the sensitivity and specificity of this technique show variable results and some suggest that brush biopsies show an increased rate of false-positive findings when used on benign-appearing oral lesions from low risk populations14. Toluidine blue is a dye that may stain nucleic acids and aid in the identification of potentially malignant tissue. This method is mainly used by experienced clinicians to aid in clinical examination and few studies have been carried out to evaluate toluidine blue as an oral cancer screening tool18. Fluorescence imaging involves exposing the tissue to blue light resulting in the autofluorescence of cellular fluorophores after excitation. Cellular alterations, such as those present in oral lesions, result in changes in the autofluorescence of the tissue19. Early studies of this technology are encouraging however more studies with larger sample sizes are required to determine if fluorescence imaging is a reliable screening tool for oral cancer18, 19. Although new technologies for oral cancer screening are currently under investigation there are currently no routine population-based screening programs in place for oral cancer2. Current recommendations for dentists and general practitioners involve visual inspection of the oral cavity and follow-up before a surgical biopsy for definitive diagnosis. These recommendations can reduce false positive rates but have not significantly improved early diagnosis rates in OSCC patients20. A blood-based biomarker used alone or in combination with other screening techniques for the early detection and monitoring of oral cancer will lead to improved outcomes for patients diagnosed with this disease. 3  1.3 Early detection of lung cancer In 2008, lung cancer accounted for 18% of overall cancer deaths and was the leading cause of cancer death in men and the second leading cause of cancer death in women worldwide21. Lung cancer is classified into two main subgroups, small-cell lung cancer and nonsmall-cell lung cancer (NSCLC). NSCLC is the most common subtype accounting for ~87% of newly diagnosed lung cancers. NSCLC is classified into four subgroups: adenocarcinoma, squamous-cell carcinoma, large-cell carcinoma, and NSCLC-not otherwise specified22. Five-year survival rates for lung cancer remain low at ~16% and, as in oral cancer, this dismal prognosis is partly due to diagnosis at late stages23. The majority of current lung cancer diagnoses occur when patients present with symptoms and seek clinical examination. Unfortunately, these symptoms usually arise at an advanced, inoperable stage24. Therefore, efforts have been made to determine the advantages of lung cancer screening tests for high-risk individuals (such as heavy smokers)22. Studies examining the use of sputum cytology and chest x-ray as lung cancer screening tools have shown that although lung cancer was detected at an earlier stage in screened individuals, there was no significant decrease in mortality rates25, 26. In 2002 the National Lung Screening Trial compared lung cancer screening with chest radiography to screening with low-dose computed tomography (CT) to see if mortality rates would decline. This group found that screening using low-dose CT showed a 20% decrease in mortality from lung cancer compared to those screened using chest radiography27. However, this technique does have risks including: a high rate of false positive results with the downstream implication of unnecessary diagnostic procedures, radiation exposure, and overdiagnosis (discovery of a cancer that would never come to light, nor cause any  4  symptoms or problems for the remainder of the person’s life either because of slow growth or because some other illness brought about death before the diagnosis of the cancer)24. Other techniques currently being examined for early lung cancer detection are autofluorescence imaging and biomarkers in sputum, exhaled breath, and blood28. Autofluorescence bronchoscopy has been used with white-light bronchoscopy to detect early stage lung cancer and preinvasive lesions in the central airways and has been shown to have an increased sensitivity compared to white-light bronchoscopy alone29. However this technique also identifies benign conditions including inflammation and is therefore associated with a low specificity30. Currently, because small lung cancers are unlikely to exfoliate detectable cancer cells, sputum cytology is not considered a useful screening tool for early lung cancer detection. However, enhanced molecular techniques have led to improvements in the diagnostic value of sputum cytology which may be used to identify individuals with an increased lung cancer risk31. Biomarkers have been identified in exhaled breath and blood however these biomarkers are not used in a clinical setting as they are still in the experimental phase and have not been validated in large studies32-34. A non-invasive test with the ability to detect early lung cancer and distinguish between benign and malignant nodules would greatly improve the prognosis of patients diagnosed with this disease.  1.4 microRNAs microRNAs (miRNAs) are endogenous, non-coding RNAs that are ~22 nucleotides in length. miRNAs control gene expression by binding to target mRNAs leading to their degradation or repression of translation35. miRNAs are transcribed in the nucleus into primary 5  miRNA transcripts called pri-miRNA which are cleaved by the RNase III enzyme Drosha into a precursor miRNA (pre-miRNA)36. The pre-miRNA is a stem loop intermediate about 60-70 nucleotides long. The export receptor Exportin-5 (Exp5) binds to pre-miRNA and exports it from the nucleus into the cytoplasm37. Once in the cytoplasm, the terminal base pairs and loop of the pre-miRNA are cleaved off by another RNase III enzyme, Dicer. This cleavage leaves a small double stranded RNA referred to as the miRNA:miRNA* duplex35. One strand of this duplex forms the mature miRNA and is incorporated into the RNA-induced silencing complex (RISC). The RISC is a ribonucleoprotein that contains the mature miRNA and an Argonaute (Argo) protein which is believed to be responsible for translational repression38. The other strand from the miRNA:miRNA* duplex, referred to as the miRNA* strand, generally accumulates at a lower level than its partner strand an is assumed to be mostly degraded however, studies suggest some miRNA* strands may have an extensive regulatory role39 (Figure 1). Once the mature miRNA is incorporated into the RISC this complex inhibits the translation of target mRNA through 1) mRNA cleavage 2) mRNA destabilization or 3) translational repression40. Endonucleolytic cleavage of the target mRNA takes place mainly in plant miRNAs and occurs when the miRNA and mRNA show perfect complementarity41. In metazoans, miRNAs imperfectly base-pair with their target mRNAs with the exception of the seed regions of the miRNA (nucleotides 2-8) which must bind perfectly with its target binding site in the 3’ UTR of the mRNA in order for optimal translational repression to take place40. Rather than mRNA cleavage, this imperfect binding leads to mRNA destabilization or translational repression. The exact mechanisms of translational repression by miRNAs are still under investigation42. Evidence exists that supports translational repression both at the initiation stage as well as post-initiation however, it is believed that the majority of translational repression 6  takes place at initiation42. Additionally, many studies have shown an inverse correlation in the levels of miRNAs and their target mRNAs suggesting that miRNAs can induce mRNA destabilization and subsequent degradation43. Further studies have shown that the mechanism of this decay is through the binding of the RISC complex followed by deadenylation of the target mRNA and degradation via an exonuclease44. Although our understanding of miRNA biology is increasing, further studies are required to fully understand the mechanisms by which miRNAs repress protein production.  7  Figure 1. miRNA biogenesis and incorporation into the RISC complex.  1.5 miRNAs and cancer Over 1500 mature miRNA sequences have been identified in humans (miRBase, Release 18.0, http://www.mirbase.org). This high number of miRNAs combined with the fact that each miRNA can target multiple mRNAs has led to the estimate that over 30% of the human genome is regulated by miRNAs45. Therefore, it is not surprising that studies have found miRNAs to be 8  involved in many cellular processes including development, cell proliferation, apoptosis, fat metabolism, and cell differentiation46-51. With such widespread regulatory functions in gene expression and key roles in cancer associated cellular processes, the roles of miRNAs in cancer are now being extensively explored52. In an early study examining the role of miRNAs in cancer, Calin et al mapped 186 miRNA genes and compared their chromosomal loci to areas commonly deleted or amplified in human tumors53. This work showed that 52% of the mapped miRNAs are located in cancerassociated genomic regions and demonstrated the potential involvement of miRNAs in many cancer types. This work has led to the classification of miRNAs as both tumor suppressors (miRNAs that target mRNAs of oncogenes and are frequently down-regulated in cancer) and oncogenes (miRNAs that target mRNAs of tumor suppressor genes and are frequently upregulated in cancer)52. In addition to this new classification miRNAs have been shown to be involved in Hanahan and Weinberg’s hallmarks of cancer54 with deregulated miRNA expression allowing cancer cells to be self-sufficient in growth signals, evade growth suppressors, resist apoptosis, obtain limitless replicative potential, induce angiogenesis, and activate invasion and metastasis55. The connections between miRNAs and cancer have led to investigations into potential clinical uses for miRNAs as prognostic, predictive, and diagnostic molecules. In 2005, a group used a bead-based flow cytometric method to examine miRNA expression in 20 human cancers56. The authors found that miRNAs are generally down-regulated in tumor compared to normal tissues and that miRNA expression signatures may be useful for the classification of poorly differentiated cancers. Since 2005, many studies have been published examining the miRNA expression profiles of tumors. While the numerous studies on tumor miRNA expression 9  have yielded promising results, and some miRNA markers are currently being tested in clinical trials, further investigation in this area is needed before miRNAs can be used in the clinic as prognostic, predictive, and diagnostic biomarkers57. Another potential clinical application for miRNAs in cancer that is under increasing investigation is their use as therapeutic targets or tools57. One feature of miRNAs that makes them attractive molecules for therapeutic use is that, compared to single genes or proteins, miRNAs regulate the expression of multiple genes and can have implications for entire pathways52. For example, the reintroduction of tumor-suppressive miRNAs miRs 15a and 16-1 induces apoptosis in leukaemic cells both in vitro and in vivo58. Knock-down of oncogenic miR21 inhibits breast tumor growth in a xenograft mouse model59. In addition to restoring tumorsuppressive miRNA expression or inhibiting oncogenic miRNA expression through anti-sense oligonucleotides, introduction of synthetic miRNAs to target oncogenic proteins has also been explored as a miRNA-based therapeutic60. Because miRNA function is so different from current targeted therapeutics, these molecules present new challenges regarding their integration into the clinic61. Nonetheless, the field of miRNAs as therapeutics for the treatment of cancer is a promising one.  1.6 Circulating miRNAs In 2008, two independent groups published studies demonstrating that miRNAs are present at detectable levels in the serum and plasma of humans62, 63. These groups also illustrated that circulating miRNAs are very stable and could withstand multiple freeze-thaw cycles, long periods of storage, extreme pH conditions, and showed resistance to RNase A digestion. One 10  group hypothesized that the stability of these circulating miRNAs can be attributed to their packing in microvesicles which are then secreted by cells into circulation63. These initial studies sparked further research to explain the stability of circulating miRNAs and to elucidate their source and function. Since 2008, exosomes, microparticles, lipoproteins, and ribonucleoprotein complexes have all been found to be carriers of circulating miRNAs64-67 (Figure 2). Exosomes are 30-100 nm vesicles formed from the inward budding of endosomes, which are formed from inward budding of the plasma membrane. When endosomes bud inward they create multivesicular bodies (endosomes containing multiple exosomes). Exosomes are released into the extracellular environment when multivesicular bodies fuse with the plasma membrane68. Studies have shown that miRNAs contained in exosomes secreted from cancer cell lines and macrophages can be transferred to recipient cells in a paracrine fashion69, 70. Microparticles are vesicles that are 0.05-1 μm in size and are released from the plasma membrane of stimulated cells71. These particles play a role in vascular inflammation and coagulation and have been shown to contain miRNAs associated with cardiovascular diseases in a composition that is significantly different from their maternal cells65. Lipoproteins are complexes that consist of a lipid core surrounded by a shell of apolipoproteins that allow the lipids to travel in the bloodstream72. A recent study has shown that high-density lipoproteins (HDLs) are loaded with miRNAs both in vitro and in vivo and that HDLs transfer miRNAs to recipient cells66. Using size-exclusion chromatography, Arroyo et al (2011) demonstrated that the majority of certain miRNA species analyzed cofractionate with Argonaute2 proteins and not with vesicles67. Currently, it is unknown whether Argonaute2-bound circulating miRNAs can interact  11  with recipient cells and it has been suggested that these miRNAs are by-products of cellular activity73. Another important result to come out of the two seminal studies on circulating miRNAs was the finding that miRNAs are differentially expressed in the serum of individuals with certain diseases including prostate cancer, lung cancer, and diabetes62, 63. This finding opened the door for researchers to evaluate the utility of circulating miRNAs as biomarkers of disease. Circulating miRNAs possess a number of characteristics that make them ideal biomarkers including: 1) their stability under various conditions, 2) their ability to be detected through noninvasive means, and 3) their ability to be easily quantified with high sensitivity and specificity using methods such as qRT-PCR74. These characteristics have led to countless studies identifying potential circulating miRNA biomarkers for conditions such as sepsis75, drug induced liver injury76, cardiovascular disease74, diabetes77, and cancer78.  12  Figure 2. Current theories on how miRNAs enter circulation via exosomes, microparticles, proteins, and lipoproteins.  1.7 Circulating miRNAs and oral cancer In 2008 Wong et al identified the up-regulation of miR-184 in tongue SCC tissue samples (N = 38 control 30 cancer) and found that plasma miR-184 levels were also increased in these patients compared to healthy controls. Furthermore, plasma miR-184 levels were significantly reduced after surgical resection of the tumor79. Similar studies in which plasma miRNA levels were shown to reflect oral cancer tissue miRNA profiles have identified miR-24 (N = 10 control 13  33 cancer)80, miR-21 (N = 20 control 50 cancer)81, 82, miR-181 (N = 12 control 39 cancer)83, miR-31 (N = 21 control 43 cancer)84, and miR-10b (N = 36 control 54 cancer)85 as up-regulated in the plasma of oral cancer patients and miR-375 as down-regulated (N = 20 control 50 cancer)81. Together, these studies suggest the potential utility of circulating miRNAs as biomarkers for oral cancer detection and possibly as predictors of recurrence and progression of oral lesions. However, there have been few studies examining global miRNA expression in patients with OSCC and therefore many potential circulating miRNA biomarkers have yet to be investigated. Also, studies using samples collected from patients with pre-invasive oral cancer (carcinoma in situ, CIS) and oral dysplasia are required in order to identify candidate circulating miRNA biomarkers for the detection of oral cancer at its earliest stages.  1.8 Circulating miRNAs and lung cancer The first study examining circulating miRNA levels in NSCLC used Solexa sequencing technology and found 91 miRNAs that were differentially expressed between the pooled sera from 11 patients with NSCLC and 21 healthy controls62. These 91 miRNAs were later tested in serum samples from 400 NSCLC cases and 220 controls leading to a panel of 10 miRNAs significantly up-regulated in NSCLC samples (miR-20a, miR-24, miR-25, miR-145, miR-152, miR-199a-5p, miR-221, miR-222, miR-223, and miR-320)86. The authors also demonstrated the utility of their panel for early lung cancer detection by accurately classifying 6 out of 7 patients with serum samples collected before (0.7-33.37 months) clinical diagnosis of NSCLC. Two studies with smaller sample sizes also identified signatures of multiple miRNAs for early lung cancer detection. Boeri et al (2011) identified a 16 ratio signature (using 15 miRNAs) that could 14  distinguish patients with NSCLC 1-2 years prior to clinical diagnosis with 80% sensitivity and 90% specificity87. While Bianchi et al (2011) identified a 34 miRNA signature that could accurately classify symptomatic and asymptomatic patients as well as benign and malignant lesions88. In addition to these studies, numerous groups have identified potential circulating miRNA biomarkers for the early detection of NSCLC (Table 1). Although the results of these studies are promising, there is little consistency in the miRNAs identified in each study. This variation could be due to multiple analytical and pre-analytical factors. These factors make comparisons between studies difficult and before the results of NSCLC circulating miRNA biomarker studies can be translated into the clinic, these factors must be identified and standardized.  Table 1. Current studies on the use of serum miRNAs as biomarkers of NSCLC.  Study  62, 86  89  Samples1  Platform  Discovery: 11 pooled NSCLC patients and 21 controls pooled. Training and Validation sets: 200 NSCLC (AC, LCC, and SCC patients stages I-IV) and 100 controls in each set.  Discovery: 11 NSCLC (AC, SCC, and LCC stages I and II) and 11 controls. Validation: 22 NSCLC (AC, SCC, and LCC stages I and II) and 31 controls.  Normalization  Candidate miRNAs  Hemolysis  Discovery: Solexa sequencing. Training and Validation: Taqman based qRT-PCR assay.  Directly normalized to serum volume  Discovery: 91 miRNA signature Validation: miR-20a, miR-24, miR25, miR-145, miR-152, miR-199a-5p, miR-221, miR-222, miR-223, and miR-320 (all upregulated).  Found differences in miRNA profiles of NSCLC serum versus blood cells  Discovery: microarray. Validation: SYBR Green based qRT-PCR.  Discovery: U6B, U6-337, 5S, and PC-HU5S. Validation: U6 and cel-miR-39.  miR-1254 and miR-5745p (up-regulated)  Not mentioned  15  Study  88  87  90  91  92  93  94  95  Samples1 Training: 25 NSCLC (AC, Stages I-IV) and 39 controls. Testing: 34 NSCLC (AC and SCC, stage I-IV) and 30 controls. All samples were collected from asymtomatic patients. Training: 19 NSCLC (AC, SCC, and other stages I-IV) and 5 control pools containing 5-7 samples. Validation: 22 NSCLC (AC, SCC, and other stages I-IV) and 10 control pools containing 5-7 samples 70 NSCLC (AC and SCC, stages I-IV) and 44 controls. Discovery: 5 NSCLC and 5 controls. Test: 88 NSCLC (AC, SCC, and other, stages I-III) and 17 controls.  Platform  Normalization  Candidate miRNAs  Hemolysis  Taqman based qRT-PCR assay  Geometric mean of miRs 197,19b, 24, 146, 15b, and 19a  34 miRNA signature  Not mentioned  Taqman based qRT-PCR assay  Ratios between all miRNAs (N = 4,950 ratios)  16 ratio signature (using 15 miRNAs)  Not mentioned  U6  miR-21 (up-regulated)  Not mentioned  U6  miR-21 (up-regulated)  Not mentioned miRNAs are also deregulated in NSCLC blood cells  Taqman based qRT-PCR assay Discovery: microarray. Test: Taqman based qRTPCR assay.  8 NSCLC samples collected at 3 time points (2 prediagnosis 1 post-diagnosis, histology and stage not available) and 6 controls  Microarray  Quantile normalization  17 miRNAs that change throughout tumor progression  Discovery: 193 NSCLC (AC and SCC, stages I-IV) and 110 controls  Taqman based qRT–PCR was performed directly in serum samples  Relative miRNA levels quantified using the formula 250-ct  miR-125b (upregulated)  Not mentioned  Training: 30 NSCLC (AC and SCC, stages I-III) and 20 controls. Test: 55 NSCLC (AC and SCC, stages I-IV) and 75 controls.  Taqman based qRT-PCR assay  The ∆Ct matrix approach  miR-15b and miR-27b (up-regulated)  Not mentioned  Synthetic celmiR-54 and celmiR-238  miR-146b, miR-221, let7a, miR-155, miR-175p, miR-27a and miR106a (down-regulated) and miR-29c (upregulated)  Not mentioned  220 NSCLC (AC, SCC, and other, stages I-II) and 220 controls.  Taqman based qRT-PCR assay  16  Study 96  97  98  99  Samples1  Platform  Normalization  Candidate miRNAs  Hemolysis  82 NSCLC (pre- and post surgery, AC and SCC, stages I-IV) and 50 controls. Discovery: 21 NSCLC and 11 controls. Test: 97 NSCLC (AC, SCC, and LCC, stages I-IV) and 20 patients with benign lung disease and 30 controls.  Taqman based qRT-PCR assay  miR-16  miR-21, miR-205, miR24, miR-30d (upregulated)  Not mentioned  Discovery: microarray Test: Taqman based qRTPCR assay  miR-1233  miR-361-3p and miR625* (down-regulated)  Not mentioned  260 NSCLC (AC, SCC, LCC, and other, stages IIIIV) and 260 controls.  Taqman based qRT-PCR assay  miR-16  miR-125b (upregulated)  Not mentioned  62 NSCLC (AC and SCC, stages I-IV) and 20 controls  SYBR Green based qRTPCR  U6  miR-126 and miR-183 (down-regulated)  Not mentioned  1. AC = Adenocarcinoma, SCC= Squamous cell carcinoma, LCC= Large cell carcinoma  1.9 Challenges associated with circulating miRNA biomarker studies The area of circulating miRNAs is a new field of research with many unknowns to be determined. The source of circulating miRNAs in both healthy individuals and those under various disease states have yet to be elucidated. Additionally, the exact mechanisms by which miRNAs are released into the bloodstream are not clear and further studies are required to determine the function (if any) of miRNAs in circulation. In addition to these functional studies, additional research on the pre-analytical and analytical features influencing serum miRNA quantification is required. One analytical area to be examined is the choice of platform for quantification. Currently, there are three commonly used methods for the quantification of circulating miRNA: microarray chip technology, quantitative reverse transcription real-time PCR (qRT17  PCR), and sequencing technology100. Sequencing is advantageous for identifying novel miRNAs and modifications associated with different disease states but has a relatively low-input and highcost compared to other methods101. However, with the recent improvements in this technology, sequencing may soon become a commonly used tool for miRNA screening102. Microarrays are high input and the most cost-efficient of the three methods but require a high concentration of input RNA and the low amount of RNA in serum and plasma samples can lead to less reliable results. qRT-PCR is more robust than microarray, has a moderate cost, and requires less starting material than both microarray and sequencing101. Considering these features, the most common study design for circulating miRNA biomarker studies is the use of microarray for initial screening to identify candidate miRNAs followed by validation by qRT-PCR101. However, more studies are required to determine the best platform to use and to standardize circulating miRNA quantification so results from various studies can be easily compared. Another important analytical parameter that requires further investigation and standardization is the selection of a suitable endogenous control for circulating miRNA studies. Unfortunately, the well defined housekeeping genes used as endogenous controls in tissue and cell line miRNA studies are not expressed or show variable expression in serum and plasma103, 104  . Potential endogenous controls have been suggested such as the small nuclear RNA U690.  However other studies have shown that U6 can be degraded in serum62 and can have highly variable expression in some samples sets105. Another commonly used endogenous control is miR-16 which has been shown by some groups to have high and stable expression across samples106, 107. However this miRNA has also been shown to be significantly differentially expressed in patients with various disease states108, 109. To compensate for the lack of reliable endogenous controls some groups have used other methods of normalization such as normalizing 18  to the geometric mean of miRNAs detected in all samples110, to total RNA62, and to synthetic exogenous miRNAs such as cel-miR-3963. However, a reliable standard method of normalization that can be applied to various circulating miRNA studies has yet to be established. The statistical methods used to determine differentially expressed miRNAs are also quite diverse among circulating miRNA studies. Many groups identify one or several differentially expressed miRNAs using Student’s t-test62, 111 or Mann-Whitney U test112, 113 and assess clinical utility by determining a cut-off and calculating sensitivity and specificity. Other groups take advantage of the high number of miRNA species present in circulation and use computational networks to identify disease-specific miRNA profiles77. The advantages of this type of analysis over studying individual miRNAs alone is that clustering algorithms allows the identification of miRNAs that are likely to have similar biological functions and representing circulating miRNAs as networks may aid in revealing the role of specific miRNAs in properties of disease at the levels of organs and organ systems114. However, when dealing with large numbers of variables the study sample size needs to be quite large to avoid over-training the data. To ensure proper evaluation of circulating miRNA data and aid in the translation of biomarkers into the clinic, researchers need to seek the advice of experienced biostatisticians during all stages of their studies115. The identification of pre-analytical factors influencing circulating miRNA expression and quantification and appropriate experimental designs are required before circulating miRNAs can be used as reliable biomarkers. The majority of current candidate circulating miRNA biomarkers have been detected using case/control studies identifying differential expression between cancer and control subjects116. These studies can be useful in identifying biomarkers however important variables should be taken into consideration including differences in: sample 19  collection and handling, demographics, medical treatment (ie. medication), environment, and lifestyle that may influence circulating miRNA expression inducing bias in study results117. These factors are also important when considering the generalizability of the study117. Additionally, to confirm the utility of potential biomarkers for early detection, prospective studies in which blood samples are collected prior to onset of disease need to be conducted. Once these factors are addressed, the utility of circulating miRNA biomarkers in the clinic can begin to be assessed.  1.10 Rationale, hypotheses and aims The development of new early detection techniques is essential to improve the outcome of patients with cancer. Blood-based biomarkers offer promising means of non-invasive detection however few molecules have been found that have a sensitivity and specificity high enough to be used in standard clinical practice. The finding that miRNAs are stably expressed in human plasma and serum and that they are differentially expressed in patients with cancer suggests the utility of these molecules as biomarkers not only for early detection, but also as means of monitoring disease progression and recurrence. However, further research into the factors that influence circulating miRNA expression and quantification is required before the full potential of these molecules as cancer biomarkers can be achieved. The hypotheses examined in this thesis are 1) that miRNAs are differentially expressed in the serum of patients with oral and lung cancer compared to non-cancer control subjects and 2) that pre-analytical factors such as sample processing, patient life-style, and fasting status can have a significant effect on serum miRNA levels. 20  Aim 1: Identification of differentially expressed miRNAs in the serum of patients with high-risk oral lesions Aim 1 of this thesis was to evaluate global miRNA expression in serum samples taken from patients with OSCC/CIS and from a group of demographically-matched control individuals with no cancer to identify miRNAs with disease-associated serum expression. Additionally, we compared global serum miRNA expression profiles in blood samples taken from individuals with disease that were collected before and after surgery, to hone the list of disease-associated serum miRNA candidates through a pairwise internally-matched comparison. This work represents the first-ever attempt to delineate serum miRNA biomarkers for head and neck malignancy and one of the first attempts to define such markers for any epithelial precancerous lesion.  Aim 2: Pre-profiling factors influencing serum microRNA levels Before circulating miRNAs can reliably be used as biomarkers of disease, the pre-profiling factors that may contribute to inconsistent miRNA quantification must be identified and the variability of circulating miRNAs in healthy individuals must be determined. Aim 2 was to examine the effect of hemolysis, fasting, and smoking on serum miRNA profiles of healthy individuals and to compare serum miRNA profiles of samples taken from the same healthy individuals over different time periods. Using the results of these comparisons we also aimed to identify a miRNA that is stably expressed across all samples and unaffected by these preprofiling variables to us as a suitable endogenous control for serum miRNA biomarker studies.  21  Aim 3: Identification of differentially expressed miRNAs in the serum of patients with early lung adenocarcinoma A number of studies have been published examining the differential expression of miRNAs in the serum of patients with NSCLC62, 88, 89. However, the candidate miRNAs identified in these studies vary considerably; possibly due to the pre-analytical factors described above or to other sources of inter-patient variability. Aim 3 was to identify NSCLC-associated serum miRNAs using a novel approach that allows us to avoid inter-patient variability by using the patient as his/her own control. Specifically, we compared the miRNA profiles of serum collected during surgery from a) pulmonary venous effluent draining the tumor vascular bed and b) systemic arterial blood from the same patient118. Given the proximity to the tumor source, effluent blood samples are expected to have the highest concentration of tumor-associated miRNAs when compared to matched systemic arterial blood samples. Furthermore, we examined the possible utility of the differentially expressed miRNAs as biomarkers for lung cancer detection by comparing the serum miRNA profiles of patients with lung adenocarcinoma to a group of non-cancer controls.  2. Materials and methods  2.1 Sample collection Aim 1 serum samples were collected from 30 patients with high-risk oral lesions (HRLs, i.e. CIS or OSCC cases) undergoing curative resection treatment between the dates of March 2010 and June 2011. Additional samples from ten of these patients were obtained six months after surgery for comparison to pre-surgery samples. For two of these patients, fresh frozen tissue 22  samples from biopsies of the lesion were also collected. Serum samples from a non-cancer control group (N=26) that were demographically matched to the above cohort for age, sex, smoking status, and ethnicity were also collected (Table 2). These samples were collected between the dates of December 2008 and March 2011. All samples used in aim 2 were collected from healthy volunteers. To determine the effects of hemolysis on serum miRNA profiles, two vials of blood were drawn from 5 volunteers. To mimic mechanical hemolysis due to improper blood collection or specimen preparation119 one vial was lysed immediately after collection by passing the blood through a 20 gauge needle several times. To assess the effects of fasting on serum miRNA profiles, blood samples were drawn from 7 healthy volunteers one hour after eating a fatty meal and again, three weeks later, after fasting overnight. Serum triglyceride levels were measured for each sample using a Triglyceride Assay Kit (Cayman Chemical Company, Ann Arbor, MI, USA). To determine the effects of smoking on serum miRNA profiles, blood samples were collected from 10 current smokers and from 10 age and sex-matched non-smokers without evidence of disease. Finally, to examine changes in serum miRNA levels over time two samples were collected from 12 volunteers over various time periods. To identify an initial list of tumor associated miRNAs for aim 3, serum samples were collected during surgery from the pulmonary venous effluent draining the tumor vascular bed and from the systemic arterial blood of the same patient (N=20, 10 from each patient). To further examine the utility of differentially expressed miRNAs as NSCLC biomarkers, we compared the serum miRNA profiles of samples collected from patients with lung adenocarcinoma (N= 26) to samples from demographically matched, high risk smokers without cancer (N=43). All subject demographics are listed in Table 3. 23  Table 2. Aim 1 patient demographics.  Oral carcinoma in situ patients  Oral squamous cell carcinoma patients  Non-cancer control patients  P-value Cancer (CIS and OSCC) vs Control  Total patients Mean age Age range  14 64 50-84  16 62 51-93  26 62 50-75  0.63a  # of males Former Smokers Current Smokers Non-Smokers Mean smoker pack years Pack year range  12 8 5 1  9 10 3 3  13 14 12 0  0.17b 0.39b  24  30  43  0.002a  Caucasian Other a Student’s t-test b Fisher’s exact test  0-60 13 1  0-156 Ethnicity: 14 2  30-80 26 0  0.24b  24  Table 3. Aim 3 patient demographics.  Total patients  Surgical Samples a 10  Mean age Age range  71.5 62-79  65 49-80  63 50-75  0.25b  # of males  4  12  25  0.46c  Former Smokers Current Smokers Never Smokers  3 4 1  9 17 0  24 19 0  0.14c  63.2 35-88  46 16.5-87.5  47 30-80  0.88b  4 1  1 4  8 (30.7) 5 (19.2) 3 (11.5) 2 (7.6) 6 (23) 2 (7.6)  10  26  Mean pack years Pack year range Stage (%): 1A 1B 2A 2B 3A Unknown Ethnicity: Caucasian a  Data missing data for some cases  b  Student’s t-test Fisher’s exact test  c  Lung P-value Adenocarcinoma Non-cancer (Adenocarcinoma Patients Controls vs Control) 26 43  43  2.2 Blood processing All blood samples were collected in SST vacutainer tubes and kept at room temperature (30 minutes) to allow clotting. Clotted samples were centrifuged at room temperature for 15 minutes at 1500 rcf. Serum was aliquoted and frozen at -80°C. All samples were processed within 2 hours of initial collection.  25  2.3 Hemoglobin concentration Serum absorbance levels were measured spectrophotometrically using a NanoDrop ND1000 spectrophotometer (NanoDrop Technologies). The hemoglobin concentration of each sample was calculated using an Allen Correction120 with the following formula:  Hemoglobin (g/L) = 1.672 A415 - 0.836 A380 - 0.836 A450  2.4 RNA purification Total RNA was extracted from 200 μL of serum using the miRNeasy Mini Kit (Qiagen) according to the manufacturer instructions, except that 1.25 μL of MS2 carrier RNA (Roche Applied Science) per 200 μL of serum was added to the QIAzol Lysis Reagent prior to RNA purification as the addition of a carrier has been shown to improve miRNA recovery121. Purified RNA was re-suspended in 50 μL of nuclease-free water and stored at -80°C prior to assaying miRNA expression. For some samples a synthetic spike-in consisting of 20 fmol of synthetic C. elegans miR-39-3p (cel-miR-39-3p; Qiagen, Toronto, ON, Canada) miRNA was added to each sample after a five minute incubation in QIAzol Lysis Reagent. For the freshly frozen oral HRL tissues, RNA was purified via a standard TRIzol (Life Technologies Inc.) protocol after microdissection by an oral pathologist.  26  2.5 miRNA quantification by qRT-PCR Due to low RNA yields in serum, the concentration of purified RNA could not be reliably measured, hence fixed volumes of eluted RNA (19.2 μL for 768 reactions) were used for miRNA expression assays as previously reported81, 84, 121. RNA was reverse transcribed using the miRCURY LNA Universal RT miRNA PCR, Polyadenylation and cDNA synthesis kit (Exiqon Inc.). cDNA was then quantified by quantitative real-time PCR (qRT-PCR) on the miRNA Ready-to-Use PCR, Human panel I and panel II with the miRCURY LNA Universal RT miRNA PCR, SYBR Green master mix according to the manufacturer’s protocol (Exiqon Inc.). For samples profiled with a synthetic spike-in, 1μl of LNA™ control cel-mir-39-3p primer (Exiqon) was added to two empty wells on each panel. Ct (cycle threshold) values and ROX reference dye normalization were calculated using Viia7 Software (Applied Biosystems). All assays were inspected for distinct melting curves – those with >1 Tm and those detected within 5 Ct’s of the negative control (Ct >35) were excluded from analysis. The lung cancer surgical samples were quantified by Exiqon qPCR Services using the same reagents, panels, and quality controls as described above. Amplification for these samples was performed in a Roche Lightcycler480.  2.6 Data analysis qRT-PCR results were exported to GenEx (MultiD Analyses AB) and normalized using inter-plate calibrators on the miRNA Ready-to-Use panels. The normalized (∆Ct) values for aim 1 samples were calculated using the mean expression value of all expressed miRNAs in a given sample (the global mean110) with the following equation: ∆Ct = global mean - Ct (assay). Aim 2 27  and aim 3 samples were normalized as described in each corresponding results section. All matched samples (samples collected from the same person) were analyzed using a fold-change analysis (2(sample 1 ∆Ct – sample 2 ∆Ct)). Mann-Whitney, ANOVA, Tukey HSD post hoc testing, and box plots were calculated and generated using STATISTICA Software (StatsoftTM, Tulsa, OK). For the aim 1 oral cancer samples, a permutation test was conducted using miRNAs that were present at detectable levels in ≥60% of samples and was used to identify miRNAs significantly deregulated between disease and control cases122. All P-values were corrected using the Benjamini and Hochberg method for multiple testing. Average-linkage hierarchical clustering was calculated and displayed using Genesis software (www.genome.tugraz.at). Receiver operating characteristic (ROC) curves and area under the curve (AUC) plots were generated using the ROCR package in R123.  3. Differential expression of miRNAs in the serum of patients with high-risk oral lesions1  3.1 Introduction There have been reports of differential miRNA expression in plasma from patients with oral squamous cell carcinoma (OSCC) or esophageal cancer80, 81, 84. Individual miRNAs examined in these studies were selected a priori, based on direct study of cancer tissues. Other data, however, suggest that miRNAs showing disease-associated expression changes in blood are  1  A version of chapter 3 has been published. MacLellan SA, Lawson J, Baik J, Guillaud M, Poh CF , and Garnis C. (2012) Differential expression of miRNAs in the serum of patients with highrisk oral lesions. Cancer Medicine 1(2): 268-274. 28  not necessarily the same ones differentially expressed in cancer tissues; tumor cells may selectively release specific miRNAs into the extracellular environment. Hence miRNA expression patterns in tumor and serum samples from the same patient might not correlate124, 125. Further, plasma miRNA expression has been shown to be more variable (as compared to serum expression)95. This inconsistency may be due to contamination of platelet miRNAs in plasma samples126, a reality that makes serum samples more attractive for blood biomarker analysis. Therefore, we evaluated global miRNA expression in serum samples taken from patients with OSCC/CIS and from a group of demographically-matched control individuals with no cancer to identify miRNAs with disease-associated serum expression.  3.2 Results  3.2.1 miRNA expression in patients with OSCC and CIS To identify differentially expressed miRNA we use qRT-PCR to quantify miRNA expression in serum from 56 subjects (30 HRL patients and 26 non-cancer controls). Of the 742 miRNAs assayed, 565 were detected in ≥1 of our samples and 58 were detected in every sample. Additionally, 502 miRNAs were detected in ≥1 HRL sample and 69 miRNAs were detected in all HRL samples. The HRL cohort consisted of 14 CIS and 16 OSCC cases. Since miRNA expression can vary based on histopathological stage127, we first asked whether these two groups had distinct serum miRNA profiles compared to controls. A one way ANOVA of all 56 samples showed significant differential expression of six miRNAs that were specific to CIS versus control comparisons. Two miRNAs were found to have significantly different expression between CIS 29  versus control and CIS versus OSCC (though not different between control and OSCC). Also, ten miRNAs were found to have differential expression between control and both CIS and OSCC but not between CIS and OSCC. And four miRNAs showed a difference specific to the OSCC group when compared to control (Table 4). Ultimately, since CIS has a high chance of progressing to OSCC, we treated CIS and OSCC as a single HRL cohort. A permutation test comparing HRL samples to the demographically-matched control samples was conducted on all 56 cases using the global-mean normalized data (∆Ct). A total of 45 miRNAs were significantly up-regulated (p<0.001) and ten miRNAs were significantly down-regulated in the HRL samples compared to controls (Table 5).  30  Table 4. Results of a one-way ANOVA followed by Tukey’s HSD post hoc to determine significantly differentially expressed miRNAs between ctrl (non-cancer control), CIS (carcinoma in situ), and OSCC (oral squamous cell carcinoma) samples. miRNA  Corrected P-value  miR-197 miR-125b miR-342-3p miR-19a miR-486-5p miR-23a let-7d* miR-26b miR-26a miR-29c miR-320a miR-142-5p miR-16 miR-29a miR-338-3p let-7b miR-17 miR-150 miR-18b miR-181a miR-223 miR-423-3p  0.0150 0.0219 0.0282 0.0331 0.0417 0.0469 0.0046 0.0300 0.0329 0.0373 0.0177 0.0040 0.0042 0.0066 0.0081 0.0093 0.0367 0.0373 0.0373 0.0480 0.0002 0.0026  Groups that are significantly different ctrl/CIS ctrl/CIS ctrl/CIS ctrl/CIS ctrl/CIS ctrl/CIS ctrl/OSCC ctrl/OSCC ctrl/OSCC ctrl/OSCC ctrl/CIS , ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, ctrl/OSCC ctrl/CIS, CIS/OSCC ctrl/CIS, CIS/OSCC  31  Table 5. miRNAs significantly (P > 0.001) up or down-regulated in cancer serum samples as determined by a permutation test. Up-regulated miRNAs hsa-miR-16 hsa-let-7b hsa-miR-320a hsa-miR-18b hsa-miR-26a hsa-miR-17 hsa-miR-19a hsa-miR-26b hsa-miR-486-5p hsa-miR-92a hsa-miR-22* hsa-miR-125b hsa-miR-205 hsa-miR-301a hsa-miR-185 hsa-miR-30e hsa-miR-20a hsa-miR-320b hsa-miR-15b* hsa-miR-126* hsa-miR-484 hsa-let-7c hsa-miR-128  Corrected P-value 1.00E-10 1.00E-10 1.00E-10 1.00E-10 1.23E-10 1.35E-10 2.11E-10 7.64E-10 8.99E-10 7.24E-09 8.24E-09 2.19E-08 5.09E-08 9.50E-08 1.33E-07 1.77E-07 2.92E-07 9.79E-07 1.06E-06 1.12E-06 1.97E-06 2.19E-06 2.22E-06  Up-regulated miRNAs hsa-miR-451 hsa-miR-7 hsa-miR-93 hsa-miR-25 hsa-miR-151-3p hsa-miR-660 hsa-miR-186 hsa-let-7a hsa-miR-195 hsa-miR-624* hsa-miR-146b-5p hsa-miR-106a hsa-miR-20b hsa-let-7i hsa-miR-21 hsa-miR-16-2* hsa-miR-93* hsa-miR-324-3p hsa-miR-27b hsa-miR-423-5p hsa-miR-106b hsa-miR-144  Corrected P-value 4.77E-06 1.11E-05 1.22E-05 1.52E-05 3.60E-05 3.98E-05 5.68E-05 6.40E-05 1.10E-04 1.30E-04 1.33E-04 1.34E-04 1.40E-04 1.83E-04 2.02E-04 2.29E-04 2.34E-04 2.71E-04 3.40E-04 6.91E-04 7.13E-04 7.67E-04  Down-regulated miRNAs hsa-miR-29a hsa-miR-223 hsa-miR-338-3p hsa-miR-142-5p hsa-let-7d* hsa-miR-150 hsa-miR-181a hsa-miR-342-3p hsa-miR-23a hsa-miR-423-3p hsa-miR-140-5p  Corrected P-value 6.55E-07 1.05E-06 1.08E-06 1.66E-06 7.54E-06 1.07E-04 1.18E-04 1.93E-04 4.96E-04 6.84E-04 9.88E-04  32  3.2.2 Serum miRNA levels pre and post surgical resection of an oral HRL  To ensure that the observed differential miRNA expression between the HRL and control groups was due to the presence of an oral lesion – and to avoid inter-patient variability – we next compared miRNA profiles for serum samples taken from matched pre- and post-surgical serum samples (N=10). Serum samples contributing to this analysis were taken immediately prior to surgery and six months following surgical resection. An average-linkage hierarchical clustering analysis on the 50 miRNAs with the highest standard deviation showed that the serum miRNA profiles clustered according to the presence/absence of disease – and not according to the source patient (Figure 3).  33  Figure 3. An average-linkage Pearson correlation hierarchical clustering analysis of the 50 miRNAs with the highest standard deviation across all pre- vs. post-surgery cases showing that disease state has more influence on serum miRNA profiles than inter-patient variability. The analysis was performed on mean centered, normalized data. 34  3.2.3. Serum miRNA biomarker candidates A fold-change analysis of miRNA expression levels showed up-regulation of 32 miRNAs and down-regulation of 41 miRNAs by ≥2-fold in ≥50% of pre-surgical cases (versus postsurgical ones). Comparing the results from both above analyses – 1) analysis of HRL versus control samples and 2) analysis of pre- versus post-surgical samples – uncovered 15 miRNAs with expression that was significantly up-regulated with disease (i.e., cancer patients in comparison #1 and pre-surgical patients in comparison #2). Five miRNAs were also found to be significantly down-regulated with disease in this comparison (Table 5). Follow-up with ROC curve analysis of these 20 candidates identified five miRNAs with AUC values >0.8. Two of five were significantly up-regulated in HRL cases compared to controls (miR-16 and let-7b) and three were down-regulated in this same comparison (miR-338-3p, miR-223, and miR-29a) (Figure 4).  35  36  37  Figure 4. (A) Expression levels of the five candidate miRNAs in the serum of oral HRL (N = 30) and non-cancer control (N = 26) patients. Scale at y-axis represents Ct values normalized to the global mean. Line inside the box: median, box: interval between the 25th and 75th percentiles, whiskers: interval between the 10th and 90th percentiles, circles: outliers. (B) ROC curve analysis conducted on the cancer versus control cases showing the five most significantly deregulated miRNAs. miR-333-3p yielded an AUC of 0.82 (95% CI: 0.71–0.94) with 80.0% specificity and 80.0% sensitivity in identifying oral HRLs, miR-29a yielded an AUC of 0.82 (95% CI: 0.70–0.93) with 76.7% specificity and 76.9% sensitivity, miR-223 yielded an AUC of 0.81 (95% CI: 0.69–0.92) with 60.0% specificity and 96.2% sensitivity, miR-16 yielded an AUC of 0.84 (95% CI: 0.73–0.94) with 93.3% specificity and 61.5% sensitivity, and let-7b yielded an AUC of 0.82 (95% CI: 0.71–0.93) with 80.0% specificity and 80.8% sensitivity.  3.2.4 Correlation between oral HRL tissue and serum miRNA profiles We also sought to determine whether serum profiles were reflective of tissue miRNA expression in two cases where matched HRL tissue specimens were available. For the tissue samples, miRNA levels of tumor tissue were compared to that of matched adjacent normal tissue and a fold-change ≥ 2-fold was considered significant. For one patient, 30 miRNAs were upregulated in both tissue and serum, no miRNAs were down-regulated in both serum and tissue, and 322 miRNAs were differentially expressed between tissue and serum. For the second patient, 39 miRNAs were up-regulated and 3 were down-regulated in both samples, while 346 were differentially expressed.  3.3 Discussion Novel biomarkers detecting oral cancer at early stages could be used to more effectively stratify patients and ultimately improve disease outcomes. Serum-expressed miRNAs have great potential as biomarkers: miRNAs are understood to be key molecular drivers affecting cancer phenotypes (thus making them worthy of direct monitoring)56 and serum samples for miRNA 38  expression are both cheaply and easily (i.e. non-invasively) obtained, making them feasible to apply in screening populations. By evaluating expression of circulating miRNAs in sera from patients with oral cancer, we identified 20 miRNAs that are significantly deregulated with oral HRLs. This work represents the first-ever global analysis of serum miRNA expression profiles in patients with any head and neck cancer or pre-cancer. By comparing the serum miRNA profiles of patients with oral cancer to profiles from non-cancer controls, we identified 55 miRNAs that were significantly deregulated with disease, demonstrating that the presence of oral HRLs can significantly alter circulating miRNA expression. Six miRNAs were specifically deregulated in the serum of patients with precancerous lesions while four miRNAs were specifically deregulated in serum from OSCC patients (Table 4). These stage-specific miRNA profiles may represent alterations in exosomallysecreted miRNAs during cancer progression. The presence of specific serum miRNAs associated with precancerous lesions also suggests utility for such miRNAs in the early detection of oral cancer; these easily tested blood markers could be applied as an adjunct to direct investigation of the oral cavity to improve disease detection. An analysis of serum miRNA levels in pre- and post-operative samples demonstrated that many circulating miRNAs are tumor-specific as the levels correspondently decrease or increase after tumor resection (Figure 3). Because miRNAs are involved in many cellular processes35, the use of these patient-matched samples allowed us to remove inter-patient variability. These results suggest that serum miRNAs could also have utility for monitoring disease recurrence, a clinical reality that contributes significantly to continuing poor oral cancer outcomes9. Moreover, because treatment (surgery, radiation) can alter the appearance of normal tissue surrounding a  39  lesion, which can delay recurrence detection, serum miRNAs may be applicable as an alternative means of post-treatment monitoring. Cross-comparison of results from the two above analyses identified 20 serum miRNAs consistently deregulated based on disease status. These miRNAs did not overlap with candidates defined by other oral cancer analyses79, 80, 84. One explanation for this could be the use of serum as the miRNA source instead of plasma, as recent reports have shown differential miRNA expression in patient-matched serum and plasma samples95, 126. Another significant result was the detection of serum miRNAs up-regulated in normal and post-surgery patients as compared to disease/matched pre-surgical samples (five of 20 miRNAs referenced above). The finding of serum miRNAs up-regulated in disease/pre-surgical samples relative to their comparators is intuitive as the given HRL still harbored by the patient could be the over-expressed serum miRNA source; absence or removal of this miRNA source would be associated with reduced expression. However, the finding of disease-associated serum miRNAs exhibiting reduced expression relative to non-disease/post-treatment states cannot be explained by this mechanism, suggesting that expression of disease-associated miRNAs in serum may be a product of factors beyond secretion of these miRNAs from diseased cells. Similar results have been reported in other cancer types81, 95. ROC analysis of the 20 miRNA candidates identified five miRNAs (miR-16, let-7b, miR338-3p, miR-223, and miR-29a) that have an AUC >0.8, suggesting that these miRNAs could have utility as biomarkers for oral cancer detection. The two miRNAs that were highly upregulated in sera from patients with oral HRLs (miR-16 and let-7b) have been reported to act as tumor suppressors (down-regulating oncogenes like RAS, BCL2) and are known to be downregulated in many cancer types128-130. Our result may be a product of selective exosome-mediated 40  release of let-7 miRNA family members into the extracellular environment, with this exclusion serving to preserve cancer phenotypes124. Indeed, selective release of tumor suppressive miRNAs by cancer cells presents a possible explanation for the up-regulation of let-7b and miR-16 in the serum of patients with OSCC/CIS. It may also explain the relative lack of correlation between our matched tissue and serum miRNA profiles reported by other groups125. Interestingly, miR-16 has been used by a number of studies as an endogenous control to normalize qRT-PCR data from both OSCC tissue and plasma samples79, 84, however our results show that it is the most significantly up-regulated miRNA in the serum of patients with oral cancer (suggesting it may not be a reliable endogenous control). The three miRNAs that were highly down-regulated in serum from patients with oral HRLs – miR-338-3p, miR-223, and miR-29a – have been shown to be down-regulated in esophageal squamous cell carcinoma tissue, OSCC cell lines, and hypopharyngeal squamous cell carcinoma tissue, respectively131-133. As mentioned above, the process by which disease-associated down-regulation of serum miRNAs is an outstanding issue requiring further investigation. We report the first-ever analysis of global miRNA expression in serum samples from patients with oral cancer or pre-cancer. By incorporating matched, pre-/post-surgical serum samples into our combined analysis, we have reduced the impact of inter-patient variability on our findings. Due to the relatively small number of samples used in this study, further validation in a larger cohort is needed to fully assess the utility of candidate miRNAs as oral cancer biomarkers. We are currently drawing samples from a large clinical trial to facilitate this validation. Current clinical practice relies on histology to determine the malignant potential of dysplastic lesions however this method is a poor predictor for low grade lesions134. Therefore, we are also accruing serum samples from patients harboring this disease stage to determine whether 41  our serum miRNA biomarker candidates are applicable to this patient population. These markers could potentially be used alone or in combination with current promising molecular markers in the literature including chromosomal alterations134 and quantitative cytology135 to predict the progression of oral dysplasias. Our results provide a strong rationale for wider evaluation of serum miRNAs as biomarkers for management of oral malignancy. They also suggest that novel processes pertaining to the relationship between disease and serum miRNA expression remain to be uncovered.  4. Pre-profiling factors influencing serum microRNA levels  4.1 Introduction Before circulating miRNAs can reliably be used as biomarkers of disease, the preprofiling factors that may contribute to inconsistent miRNA quantification must be identified and the variability of circulating miRNAs in healthy individuals must be determined. Recent studies have shown that the majority of miRNAs found in serum and plasma are also present in blood cells and that hemolysis during sample collection or processing may affect the miRNA profile of the sample62, 136, 137. In particular, circulating miRs 451, 16, 24, 15b, 223, 486, 150 and 92a have all been shown to be inconsistently expressed in healthy individuals due to their presence in blood cells126, 138. However, the effects of mechanical hemolysis on whole micronome serum quantification have not been extensively studied. Additionally, it has been suggested that lipid intake before blood draw may interfere with miRNA extraction leading to a variable profile139. To date there have been no studies examining the effects of fasting status on miRNA expression profiles. Lifestyle may also affect serum miRNA expression. Because we (along with others62, 88, 42  93  ) are interested in the utility of circulating miRNAs as lung cancer biomarkers and because  ~85% of lung cancers in men and 45% in women are caused by smoking140, there is a need to examine the effects of smoking on circulating miRNA profiles. One study has shown that serum levels of miR-625* are significantly lower in non-small cell lung cancer patients who smoke compared to those who do not smoke97. However, there have been no studies comparing the serum miRNA levels of healthy smokers to healthy non-smokers. Other unknown factors such as changes in hormone levels or lifestyle changes may influence the serum miRNA profiles of healthy individuals leading to a change in these profiles over time. In this study, we examined the effect of hemolysis, fasting, and smoking on serum miRNA profiles of healthy individuals and compared serum miRNA profiles of samples taken from healthy individuals over different time periods. Our results contribute to a better understanding of the pre-analytical factors that may influence serum miRNA profiles.  4.2 Results  4.2.1 The effects of hemolysis on serum miRNA levels To determine the effects of hemolysis on serum miRNA profiles, we measured the serum miRNA profiles of lysed and unlysed serum samples from ten healthy volunteers. Hemoglobin levels of all samples were measured spectrophotometrically, all unlysed samples had a hemoglobin concentration below 1 g/L and all lysed samples had a hemoglobin concentration above 1.5 g/L (Table 6).  43  Table 6. Hemoglobin concentrations of matched lysed and unlysed serum samples.  Sample ID  Hemoglobin Concentration (g/L) Unlysed  Hemoglobin Concentration (g/L) Lysed  11  0.51  5.52  12  0.48  12.17  14  0.55  6.20  15  0.56  2.20  16  0.42  1.54  17  0.74  10.91  18  0.69  10.48  19  0.75  24.28  20  0.37  12.35  21  0.70  22.49  miR-122 was used to normalize the data because it showed little variability between samples collected from the same person and has been shown by others to be unaffected by hemolysis126, 138. Additionally, a synthetic spike-in (cel-miR-39-3p) was added and measured in 14 of the 20 samples and the fold-change results of these samples normalized to miR-122 were not significantly different from the results of the spike-in normalized data. However, because the levels of miR-122 were highly variable from person to person, miR-122 normalized data was only used to make intra-individual comparisons. Out of the 742 miRNAs profiled 109 miRNAs were detected in all 20 samples, 181 miRNAs were detected in all 10 hemolysed samples, and 116 miRNAs were detected in all 10 non-hemolysed samples. 36 miRNAs were detected in at least 5 of the 10 hemolysed samples at a Ct below 34 but were not detected in any of the matched non-hemolysed samples.  44  A cut-off of 2-fold is commonly used to identify significant changes in circulating miRNA levels86, 141 however, to increase our stringency and identify only the most significantly deregulated miRNAs for exclusion of further analyses a 3-fold cut-off was applied to our foldchange analysis of matched samples. This analysis identified 4 miRNAs that were up-regulated by at least 3-fold in all hemolysed samples and 144 miRNAs that were up-regulated by at least 3fold in 40% of hemolysed samples compared to unlysed samples (Table 7). No miRNAs were down-regulated by at least 3-fold in more than 1 hemolysed sample. The cel-miR-39-3p spike-in was not significantly differentially expressed between hemolysed and non-hemolysed samples suggesting that the difference in miRNA levels is not due to the presence of inhibitors in the samples. Because clinical samples often contain some hemolysis142, we have identified a list of miRNAs significantly affected by mechanical hemolysis to exclude from serum miRNA biomarker studies. For our purposes, we decided to exclude miRNAs up-regulated by at least 3fold in at least 50% of samples or detected in 80% of hemolysed samples and not detected in matched unlysed samples leaving a final list of 130 miRNAs to be excluded due to the influence of hemolysis (Table 7). To measure the correlation between hemoglobin concentration and serum miRNA levels, the raw data was normalized to miR-99a (rationale described below). Of the miRNAs detected in at least 80% of the hemolysed samples or 80% of the non-hemolysed samples, 150 miRNAs were significantly (corrected P-value < 0.05) correlated with hemoglobin concentration (Table 7). A Pearson correlation cluster analysis of miRNAs show that the samples cluster according to hemoglobin concentration and not according to sample source (Figure 5).  45  Table 7. List of 144 miRNAs most affected by mechanical hemolysis.  miRNA  # of cases  miR-15b* miR-451 miR-20b miR-144 miR-92a miR-19b miR-140-3p miR-324-3p miR-590-5p miR-15a miR-19a miR-660 miR-25 miR-16 miR-16-2* miR-32 miR-363 miR-101 miR-192 miR-210 miR-500a miR-194 miR-423-5p miR-106a miR-93 miR-20a miR-185 miR-425 miR-29b miR-106b miR-222 miR-598 miR-320a miR-652 miR-18a miR-17  10 10 10 10 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8  Median Foldchange 15.34 18.03 17.18 29.63 13.98 15.43 14.95 14.69 10.57 15.28 13.95 16.63 21.45 22.16 15.68 26.37 22.38 16.15 9.03 18.17 19.11 8.86 9.29 11.60 13.85 15.29 16.99 12.87 17.50 13.57 5.87 6.22 11.15 11.32 9.94 11.29  miRNA  # of cases  miR-335 miR-140-5p miR-21 miR-151-3p miR-26b miR-378 miR-148a miR-320b miR-30e miR-130b miR-502-3p miR-501-5p miR-454 miR-196b miR-532-5p miR-215 let-7d let-7d* miR-191 miR-107 miR-301a miR-130a miR-584 let-7g miR-345 miR-7 miR-181a miR-374a miR-421 miR-331-3p miR-324-5p miR-532-3p miR-30d miR-628-3p miR-629 miR-133b  8 8 8 8 8 8 8 8 8 8 8 8 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 6  Median Foldchange 5.62 7.70 6.89 5.71 9.09 8.65 9.49 9.68 10.30 7.48 12.08 9.91 7.50 5.34 10.45 13.16 4.02 5.99 5.69 7.53 6.66 13.46 6.61 8.56 5.37 12.53 4.83 5.41 8.25 6.89 8.63 5.90 11.50 6.63 19.62 3.94  miRNA  # of cases  miR-376c miR-423-3p miR-625* miR-942 miR-132 miR-103-2* miR-186 miR-30c miR-152 miR-182 miR-338-3p miR-27a miR-146a miR-221 miR-223* miR-151-5p miR-29a miR-24 miR-574-3p miR-223 miR-1979 miR-126 miR-143 miR-26a miR-195 miR-1974 miR-501-3p miR-30e* let-7a miR-146b-5p miR-27b miR-382 miR-33a miR-23a miR-505 let-7b*  6 6 6 6 6 6 6 6 6 6 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 4 4 4 4 4 4  Median Foldchange 4.43 7.03 5.36 6.13 5.33 11.36 14.84 3.91 5.33 4.40 3.52 2.75 2.97 3.31 3.11 4.79 3.06 3.51 3.20 3.14 3.35 4.13 3.02 3.43 3.14 3.06 3.31 5.39 3.75 3.63 2.43 2.65 3.25 2.47 2.52 2.38 46  miRNA  # of cases  let-7b miR-424 miR-93* miR-18b miR-29c miR-148b miR-22 let-7i miR-486-5p miR-328 miR-15b miR-425*  8 8 8 8 8 8 8 8 8 8 8 8  Median Foldchange 11.99 9.65 8.64 13.90 11.27 11.91 14.27 13.58 17.85 4.52 6.64 4.92  miRNA  # of cases  miR-150 miR-126* let-7c miR-103 miR-142-5p miR-144* miR-2110 miR-484 miR-30b miR-374b miR-199a-3p miR-190  6 6 6 6 6 6 6 6 6 6 6 6  Median Foldchange 3.66 3.49 4.01 5.63 4.47 4.39 5.79 14.05 3.44 5.65 3.97 3.97  miRNA  # of cases  miR-28-3p miR-376a miR-142-3p miR-199a-5p miR-582-5p miR-339-3p let-7f miR-18a* miR-550a* miR-92b miR-486-3p  4 4 4 4 4 4 4 4 4 4 4  Median Foldchange 2.23 3.69 1.89 2.86 3.19 3.01 1.90 11.20 10.87 10.62 33.51  47  Figure 5. An average-linkage Pearson correlation cluster analysis of the serum miRNA profiles of matched unlysed (yellow) and lysed (blue) samples collected from 10 healthy individuals. Only miRNAs detected in at least 80% of lysed samples or 80% of unlysed samples are included. The analysis was performed on mean centered, normalized data. 48  4.2.2 Endogenous controls for serum miRNA studies Currently, there are no standard endogenous controls for serum miRNA studies143. The varying levels of miRNAs observed in samples with high hemoglobin concentrations suggest that some current methods of normalization may be unreliable when samples exhibit hemolysis. Our data suggest that normalizing to liver specific miR-122 may be appropriate for matched samples as it is not affected by hemolysis and shows little variation in samples collected from the same person within a short period of time. However, because this miRNA is inconsistently expressed from person to person and has been shown to be deregulated under certain pathologies144, 145 it is not a suitable endogenous control for inter-individual comparisons. To identify a suitable endogenous control for serum miRNA studies, we examined the profiles of 29 serum samples from healthy individuals, 7 of which had a hemoglobin concentration above 1.5 g/L. Synthetic miR-cel-39-3p was added to all samples prior to extraction and the levels were measured along with all other RNAs contained on the Exiqon platform. The raw Ct values of all 742 miRNAs as well as U6 and the global mean110 were converted to a linear scale and compared to the linear raw values of the synthetic miRNA. After including only miRNAs present at a detectable level in all samples along with U6 and the global mean, the levels of miR-99a showed the highest correlation (highest R-value) with miR-cel-393p levels. When all samples were normalized to cel-miR-39-3p, miR-99a also had the lowest standard deviation across all samples compared to the other potential controls. When hemolysed samples were removed, the correlation and standard deviation values for miR-99a remained similar, however when these samples are removed, 9 miRNAs (miR-29b, miR-144, miR-29c, miR-590-5p, miR-29a, miR-103-2*, miR-27a, miR-32, and miR-101) had a higher R-value than miR-99a. When compared to three commonly used endogenous controls (miR-16, U6, and the 49  global mean) miR-99a has a higher correlation to cel-miR-39 and a lower standard deviation when normalized to cel-miR-39-3p than all 3 controls, both with and without hemolysed samples (Table 8).  Table 8. The correlation of potential endogenous controls to spiked-in synthetic miR-cel-39-3p levels and the standard deviation of the controls when normalized to cel-miR-39-3p with and without samples exhibiting high hemoglobin concentrations. Correlation to cel-miR-39-3p (R) 0.66 -0.10 0.33 0.30  Standard Deviation when Normalized to cel-miR-39-3p 0.74 1.21 1.37 2.10  Excluding samples with hemolysis: Correlation to Endogenous control cel-miR-39-3p (R) miR-99a 0.65 b U6 -0.10 Global Mean 0.47 miR-16 0.44  Standard Deviation when Normalized to cel-miR-39-3p 0.71 1.26 0.78 0.95  Endogenous control miR-99a U6a Global Mean miR-16  a  U6 was detected in 17 of the 29 samples U6 was detected in 13 of the 22 samples  b  4.2.3 The effects of fasting on serum miRNA profiles of healthy individuals It has been suggested that fasting status may alter serum miRNA profiles due to lipids interfering with the RNA extraction139 or to miRNAs from food entering circulation146. However, to our knowledge, this effect has not yet been tested. To assess the effects of fasting on serum miRNA profiles we collected blood samples from 7 healthy volunteers one hour after eating a fatty meal and again three weeks later, after fasting overnight. Serum triglyceride levels were  50  measured enzymatically and were higher in the non-fasting samples (Table 9). A comparison of the total number of miRNAs detected in each sample showed that, after the exclusion of miRNAs significantly affected by hemolysis, the total miRNA count was slightly higher in the non-fasting samples compared to the fasting samples although not significant (P = 0.27, Figure 6). Pearson correlation coefficients between serum triglyceride concentration and miRNA levels showed no miRNAs were significantly correlated with serum triglyceride levels. A fold-change analysis between matched samples normalized to miR-122 showed that no miRNAs were significantly (by at least 3-fold in at least 50% of cases) differentially expressed between the fasting and non-fasting samples. No miRNAs were differentially detected in at least 4 of the 7 fasting or non-fasting samples and not detected in the corresponding matched sample. A MannWhitney U test of miRNAs normalized to miR-99a and expressed in at least 75% of fasting or 75% of non-fasting samples showed that no miRNAs were significantly (P < 0.05) differentially expressed between the two groups (data not shown).  Table 9. Triglyceride concentration of fasting and non-fasting serum samples. Sample 1 2 3 4 5 6 7  Triglyceride concentration (mg/dl) Non-Fasting 202.8 152.1 148.5 93.3 176.4 72.4 53.5  Triglyceride concentration (mg/dl) Fasting 120.6 96.2 52.1 30.4 102.4 57.8 45.7  51  Figure 6. The total number of miRNAs (out of 742 profiled) present at detectable levels in fasting and non-fasting samples. P-value = 0.27, Mann-Whitney U test.  4.2.4 The effects of smoking on serum miRNA profiles To examine the effects of tobacco smoke on serum miRNA expression we compared the profiles of 10 current smokers to 10 age and sex-matched controls (Table 10). A Mann-Whitney U test conducted on miRNAs normalized to miR-99a and detected in at least 80% of smokers or 80% of non-smokers showed that no miRNAs were significantly (corrected P-value < 0.05) differentially expressed. However, miR-128 showed the lowest uncorrected P-value (0.004) and is the only miRNA with an average fold-change greater than 3. This miRNA was detected in 7 non-smoker cases and 9 smoker cases and was slightly up-regulated in smokers (Figure 7). No miRNAs were present in only smoking samples or only non-smoking samples. 52  Table 10. Patient demographics for smoking/non-smoking samples.  Males Females Median age Age Range Median Pack Year  Current-Smokers  Non-Smokers  3 7 60.5 52-67 46.3  3 7 52 40-64 0  Figure 7. Levels of miR-128 in the serum of non-smokers and current smokers. Uncorrected Pvalue = 0.004, Corrected P-value = 0.78, Mann-Whitney U test.  53  4.2.5 Changes in serum miRNA levels over time To assess changes in serum miRNA levels in healthy individuals over time we compared the profiles of two samples collected from the same individual over varying time periods. Pearson correlation coefficients of the levels of miRNAs detected in at least 19 of 24 samples showed a strong correlation between samples collected over different time periods from the same person (Table 11). A fold-change analysis showed no miRNAs were frequently significantly deregulated between matched samples (up or down-regulated by at least 3-fold in at least 50% of cases). Nine miRNAs (miR-888, miR-454, miR-10a, miR-181c, miR-1909, miR-20a*, miR-484, miR-501-5p, and miR-622) were detected in at least 50% of cases and not detected in the corresponding matched control. Of these nine miRNAs, three (miR-454, miR-484, and miR-5015p) are also in the list of 130 miRNAs significantly affected by hemolysis and therefore the difference in miRNA levels could be due to differences in the amount of hemolysis in each sample.  54  Table 11. Pearson correlation coefficients for miRNA profiles of serum collected at two different time points from healthy individuals.  Sample 1 2 3 4 5 6 7 8 9 10 11 12  Age at first blood draw 60 63 54 75 67 55 64 63 69 58 60 53  Sex M F F M M M M M M M M F  Correlation Coefficient 0.89 0.92 0.92 0.93 0.94 0.94 0.94 0.95 0.95 0.96 0.96 0.96  Time Between Sample Collection 13 months 7 days 15 Months 7 days 17 months 3 days 11 months 9 days 7 months 10 days 5 Months 3 days 13 months 12 months 19 days 2 months 14 days 6 months 7 days 10 months 11 months 22 days  4.3 Discussion An ideal biomarker can be measured non-invasively and is highly sensitive and specific to the disease state of interest. In order for a biomarker to have a high positive predictive value it should not be significantly influenced by technical variables or pathologic conditions unrelated to the disease being investigated. In this study we examined the effects of a range of technical and individual factors on serum miRNA levels. Our results have identified a list of miRNAs that are significantly affected by these factors and are therefore likely to have very limited utility as biomarkers of disease. We have contributed to previous studies examining the role of blood cell contaminants in serum miRNA quantification126, 137, 138. While miRNA profiles of blood cells and blood microvesicles have been previously identified137, to our knowledge, this is the first study 55  specifically examining the effect of mechanical hemolysis on the quantification of 742 miRNAs in the serum of healthy individuals. Other studies examining the effects of hemolysis on circulating miRNA profiles have done so by profiling blood cells directly or by adding hemolysate back into already separated serum or plasma samples126. However, Dimeski (2004) demonstrated that samples prepared in this way can have a different effect on analytes than samples prepared by mimicking mechanical hemolysis119. By mimicking hemolysis caused by improper blood collection or preparation we have identified a list of 63 miRNAs that are most likely to be affected by hemolysis in a clinical setting. Currently, researchers examining the role of hemolysis on plasma and serum miRNA profiles have suggested excluding hemolysed samples from studies147. This method may be useful in identifying biomarkers because it allows the measurement of all miRNAs while avoiding the effects of hemolysis. However, from a clinical standpoint, this method is impractical because the most common technique for identifying hemolysed samples in the clinic is by visual inspection148 and our results as well as other studies147 have shown that minor (not visually apparent) levels of hemolysis can significantly alter miRNA levels. Additionally, for samples with obvious hemolysis, asking patients to return for multiple blood draws requires further staff time and supplies and can delay diagnosis leading to increased stress on the patient149. Therefore, we propose that rather than excluding hemolysed samples from biomarker studies, researchers should eliminate miRNAs significantly affected by hemolysis when identifying potential biomarkers. Several groups have addressed the importance of assessing analytical parameters for serum miRNA studies especially with regards to comparing results from multiple groups101, 126. In this study we’ve demonstrated a method for normalization that excludes miRNAs affected by hemolysis and is based on the levels of a synthetic spike-in. In cases with samples that may be 56  compromised by hemolysis, this method is more appropriate than some commonly used normalization methods. For example, normalizing to the geometric mean of miRNAs expressed in all samples110 can produce skewed results if a sample is contaminated with miRNAs from blood cells. This occurs because more highly expressed serum miRNAs, which are used for global mean normalization, are also expressed in blood cells and will be present at higher levels in hemolysed samples138. Therefore, samples exhibiting hemolysis will appear to have high RNA input due to an overall lower normalization value and miRNAs that are not expressed in blood cells will seem to be down-regulated in that sample. Another commonly used method for normalization is the use of miR-16 as an endogenous control79. However, we have demonstrated that the levels of serum miR-16 are significantly increased in samples with high hemoglobin concentrations suggesting that this method is also inappropriate in cases where some samples may be contaminated with blood cell miRNAs. Finally, another commonly used method for miRNA normalization is the use of U6 as an endogenous control. U6 is a nuclear RNA and has been used to normalize tissue and cell miRNA expression150. However, because U6 is localized in the nucleus it should not be present at consistent, high levels in the non-cellular components of blood143. Indeed, of the 60 samples analyzed in this study U6 is only present at a detectable level in 31 samples suggesting U6 is not a suitable endogenous control for serum miRNA studies. We propose that miR-99a should be used for sample normalization instead of these commonly used endogenous controls due to its correlation with synthetic spike-in levels, its low standard deviation in spike-in normalized samples, and because it is unaffected by hemolysis (Table 7). A recent report has shown that plant miRNAs are present in the sera of humans and that these miRNAs are acquired through food intake146. This finding suggests that miRNAs acquired through food intake would show variable expression over time and could be affected by fasting 57  status. Furthermore, fasting status alters the amount of lipids in the blood which could interfere with RNA extraction leading to variable miRNA levels139. Here we show that there is no significant difference in the total number of miRNAs detected in fasting versus non-fasting samples demonstrating that the presence of lipids in the blood does not lead to a loss of less abundant miRNAs during extraction. Furthermore, no miRNAs examined in our assay were significantly differentially expressed (P-value < 0.05) between fasting and non-fasting samples indicating that fasting status will not interfere with serum miRNA biomarker discovery in subjects with normal miRNA physiology. However, our study only examines short term dietary effects on circulating miRNAs and experiments with long-term, controlled diets need to be carried out in order to test the overall effects of diet on serum miRNA profiles151. Additionally, because lipoproteins have been shown to carry miRNAs in circulation66 the correlation between high density and low density lipoproteins and serum miRNAs should also be examined in future studies. As smoking plays an important role in the development of lung, oral, and other cancers and must be considered when identifying potential biomarkers for these diseases, we have also examined the influence of smoking on serum miRNA profiles. Because no miRNAs were significantly differentially expressed between smokers and non-smokers after correction for multiple testing, our results suggest that smoking status will not interfere with serum miRNA biomarker studies. However, a recent study using 6-month cigarette smoking in mice found that the levels of miR-128 were significantly changed in the lung tissue and plasma of exposed mice suggesting a role for this miRNA in the cellular response to cigarette smoke exposure152. We have found that although miR-128 was not significantly (P-value < 0.05) deregulated in the serum of smokers, this miRNA was detected in more current smoker samples than non-smoker 58  samples and was the only miRNA with an average fold-change over 3-fold higher in smoking samples compared to non-smoking samples. Therefore, further studies are required to determine if miR-128 is significantly affected by smoking status. Roth et al determined that miR-625* is down-regulated in the serum of NSCLC patients who smoke compared to those who do not smoke97. We did not see a significant difference in miR-625* between healthy smokers and nonsmokers, however this miRNA was only detected in 7 of 10 non-smoking samples and only 5 of 10 smoking samples. Therefore, using more sensitive profiling techniques or a larger sample set, miR-625* could be found to be significantly down-regulated in the serum of healthy smokers. Taking the results of both studies into account, the role of smoke exposure should be considered when determining the possible utility of miR-128 and miR-625* as biomarkers of disease. When identifying biomarkers for disease detection and/or monitoring, an important aspect to consider is the possibility of changes in biomarker levels over time in the healthy population. In order to evaluate serum miRNA variability in healthy individuals, we measured the levels of miRNAs in paired serum samples taken from 12 individuals over varying time periods. Pearson correlation coefficients of the matched samples show that the samples with the 3 lowest R-values are samples with the longest time between draws indicating that serum miRNA levels may show a slight change over longer time periods. However, the other 9 samples show no pattern between correlation coefficients and time between sample collections. This lack of a pattern and the fact that the matched samples with the lowest R-value (0.89) are still highly correlated suggest that, up to 17 months, overall serum miRNA levels show little variability in healthy individuals. Further, long-term studies are required to determine changes in serum miRNAs over several years. A fold-change analysis showed no individual miRNA was consistently deregulated in these matched samples implying that variability in serum miRNAs 59  over time should not interfere with biomarker studies examining abundant serum miRNAs. Among the less abundant miRNAs, 9 were differentially detected in the matched samples. This result demonstrates that these less abundant miRNAs are not suitable biomarkers due to their inconsistent expression. Although this study has limitations such as small sample sizes, the results presented here identify factors that should be taken into consideration when selecting endogenous controls and biomarker candidates. In order for serum miRNA biomarkers to eventually be implemented in a clinical setting, pre-analytical and analytical variables affecting serum miRNA profiles should be examined and standardized. In addition to the variables described in this study there are other areas requiring further examination. Studies have shown that pregnancy can influence serum miRNA levels153 suggesting that hormonal activity may also interfere with serum miRNA biomarker identification. Other factors requiring further investigation include the affect of benign conditions such as inflammation and diurnal variation on serum miRNA profiles of healthy individuals as well as analytical factors such as the best platform and statistical tests to use for the detection and determination of differentially expressed miRNAs. Once these variables are well defined, the complete potential of circulating miRNAs as biomarkers can be fully explored.  60  5. Identification of differentially expressed miRNAs in the serum of patients with early lung adenocarcinoma  5.1 Introduction A number of studies have been published examining the differential expression of miRNAs in the serum of patients with NSCLC62, 88, 89. However, the candidate miRNAs identified in these studies vary considerably. Some of this variation may be attributed to the analytical and pre-analytical factors described in chapter 4 while other possible sources of variation such as hormonal activity and long term differences in diets have yet to be examined. To avoid these inter-patient variables, we used a novel approach to identify tumour-associated serum miRNAs and examined their utility as non-invasive biomarkers by quantifying their levels in patients with lung adenocarcinoma and a group of non-cancer controls.  5.2 Results  5.2.1 Identification of differentially expressed miRNAs in patient-matched samples To identify an initial list of tumor-associated miRNAs we compared the miRNA profiles of serum collected during surgery from a) pulmonary venous effluent draining the tumor vascular bed and b) systemic arterial blood from the same patient. A total of 63 miRNAs were detected in at least 50% of tumor associated samples and not detected in the matched systemic case and 3 miRNAs were detected in 50% of systemic samples and not detected in the matched tumor associated sample (Table 12). A fold-change analysis identified miR-375 as up-regulated 61  by at least 3-fold in all tumor associated (pulmonary venous) samples compared to systemic blood samples. In total, 64 miRNAs were up-regulated and no miRNAs were down-regulated by at least 3-fold in at least 50% of samples (Table 12). In chapter 4 we identified a list of miRNAs significantly affected by hemolysis. To identify a final list of candidate miRNA biomarkers, we removed miRNAs significantly (by 3-fold in at least 50% of samples) affected by hemolysis. After removal of miRNAs significantly affected by hemolysis, 35 miRNAs were up-regulated by at least 3-fold in at least 50% of samples (Table 12). In addition to fold-change values, the overall numbers of miRNAs detected were higher in tumor associated samples compared to systemic arterial samples both before and after the removal of hemolysis associated miRNAs with the exception of sample 95 (Figure 8).  62  Table 12. Results of a fold-change analysis of pulmonary venous and systemic arterial lung cancer samples.  microRNA  # of samples up-regulated  Median Foldchange  Detected in ≥ 50% of tumor associated samples  # of cases  hsa-miR-375  10  13.23  hsa-miR-138  9  hsa-miR-604  5  hsa-miR-141  9  26.75  hsa-miR-216a  9  hsa-miR-20b*  5  hsa-miR-200c  9  22.46  hsa-miR-29c*  9  hsa-miR-509-3p  5  hsa-miR-125b  9  19.85  hsa-miR-100  8  hsa-miR-34a  9  12.96  hsa-miR-187  8  hsa-miR-497  9  8.43  hsa-miR-195*  8  hsa-miR-145  9  7.65  hsa-miR-642  8  hsa-miR-200b  8  14.29  hsa-miR-195  7  hsa-miR-30a*  8  7.32  hsa-miR-203  7  hsa-miR-10a  8  6.86  hsa-miR-30d*  7  ǂhsa-miR-146b-5p  8  4.86  hsa-miR-424*  7  ǂhsa-miR-133b  8  4.18  hsa-miR-663  7  ǂhsa-miR-143  7  10.02  hsa-miR-99b*  7  hsa-miR-204  7  6.47  hsa-miR-1201  6  ǂhsa-miR-29b  7  4.75  hsa-miR-125a-3p  6  hsa-miR-99a  7  4.6  hsa-miR-125b-2*  6  ǂhsa-miR-532-3p  7  3.98  hsa-miR-1260  6  ǂhsa-miR-29a  6  5.17  hsa-miR-135a  6  hsa-miR-95  6  4.89  hsa-miR-148a*  6  ǂhsa-miR-29c  6  4.88  hsa-miR-200a  6  hsa-miR-133a  6  4.77  hsa-miR-200b*  6  hsa-miR-199b-5p  6  4.34  hsa-miR-211  6  hsa-miR-193b  6  4.18  hsa-miR-214  6  ǂhsa-miR-30e*  6  3.56  hsa-miR-217  6  hsa-let-7f-2*  6  3.45  hsa-miR-23a*  6  hsa-miR-1  6  3.41  hsa-miR-23b*  6  ǂhsa-miR-423-3p  6  3.39  hsa-miR-362-3p  6  ǂhsa-miR-222  6  3.23  hsa-miR-381  6  hsa-miR-99a*  5  6.79  hsa-miR-429  6  hsa-miR-218  5  6.13  hsa-miR-449a  6  Detected in # of ≥ 50% of cases systemic samples  63  microRNA  # of samples up-regulated  Median Foldchange  Detected in ≥ 50% of tumor associated samples  # of cases  hsa-miR-30a  5  5.33  hsa-miR-511  6  hsa-miR-361-3p  5  4.68  hsa-miR-548b-3p  6  ǂhsa-miR-338-3p  5  4.62  hsa-miR-708  6  hsa-miR-18a*  5  4.31  hsa-miR-92a-1*  6  hsa-miR-181b  5  4.1  hsa-miR-101*  5  hsa-miR-1972  5  3.85  hsa-miR-124  5  hsa-miR-551a  5  3.84  hsa-miR-1271  5  ǂhsa-miR-660  5  3.65  hsa-miR-135b  5  hsa-miR-99b  5  3.5  hsa-miR-1468  5  hsa-miR-337-3p  5  3.33  hsa-miR-146b-3p  5  hsa-miR-28-3p  5  3.32  hsa-miR-149*  5  ǂhsa-miR-629  5  3.31  hsa-miR-181a*  5  hsa-miR-769-5p  5  3.27  hsa-miR-183  5  ǂhsa-miR-574-3p  5  3.22  hsa-miR-185*  5  hsa-miR-365  5  3.21  hsa-miR-193a-3p  5  ǂhsa-miR-328  5  3.18  hsa-miR-196b*  5  hsa-miR-342-3p  5  3.17  hsa-miR-26a-1*  5  ǂhsa-miR-532-5p  5  3.16  hsa-miR-299-5p  5  ǂhsa-miR-7  5  3.15  hsa-miR-331-5p  5  ǂhsa-miR-152  5  3.13  hsa-miR-33a*  5  ǂhsa-miR-940  5  3.1  hsa-miR-342-5p  5  ǂhsa-miR-132  5  3.09  hsa-miR-34b*  5  ǂhsa-miR-502-3p  5  3.03  hsa-miR-362-5p  5  ǂhsa-miR-140-3p  5  2.98  hsa-miR-489  5  ǂhsa-miR-501-3p  5  2.88  hsa-miR-519a  5  ǂhsa-miR-324-3p  5  2.86  hsa-miR-520c-3p  5  ǂhsa-miR-148a  5  2.77  hsa-miR-579  5  ǂhsa-miR-21  5  2.72  hsa-miR-589*  5  hsa-miR-17*  5  2.71  hsa-miR-629*  5  hsa-miR-326  5  2.66  hsa-miR-663b  5  ǂhsa-miR-486-5p  5  2.63  hsa-miR-675b  5  ǂhsa-miR-320b  5  2.56  hsa-miR-760  5  hsa-miR-22*  5  2.53  hsa-miR-98  5  ǂhsa-miR-146a 5 ǂ Significantly affected by hemolysis  2.19 64  Figure 8. The number of miRNAs detected in pulmonary venous (black) and systemic arterial (light grey) blood samples after removal of miRNAs significantly affected by hemolysis. P = 0.001, Mann-Whitney U test.  5.2.2 Comparison of miRNA profiles from patients with lung adenocarcinoma and noncancer controls To examine the utility of differentially expressed miRNAs as non-invasive biomarkers we compared the serum miRNA profiles of 26 patients with lung adenocarcinoma a group of demographically matched non-cancer controls (Table 3). In total, 81 miRNAs were detected in all 26 cancer samples and 94 miRNAs were detected in all control samples. An uncentered Pearson correlation cluster showed no pattern in the clustering of cancer and control samples suggesting the presence of a tumour does not affect the overall serum miRNA profiles of patients (Figure 9). A Mann-Whitney U test conducted on miRNAs normalized to miR-99a and detected in at least 80% of adenocarcinoma samples or 80% of controls showed that 22 miRNAs were 65  significantly (P-value < 0.05) differentially expressed. However, after correction for multiple testing all P-values were above 0.1 (Table 13). Interestingly, of the miRNAs with significant uncorrected P-values, only 3 were up-regulated and 19 were down-regulated in cancer samples compared to controls (Table 13). Additionally, 12 of these 22 miRNAs are listed in chapter 4 as significantly affected by hemolysis.  66  Figure 9. An average-linkage Pearson correlation cluster analysis of the serum miRNA profiles of lung adenocarcinoma patients (yellow) and non-cancer controls (blue). Only miRNAs detected in at least 80% of cancer or 80% of control samples are included. The analysis was performed on mean centered, normalized data. 67  Table 13. miRNAs significantly (uncorrected P-value < 0.05) up or down-regulated in lung adenocarcinoma serum samples compared to controls. Mann-Whitney U test, P-value corrected using Benjamini-Hochberg correction. miRNA Down-regulated hsa-miR-199a-5p ǂhsa-miR-191 hsa-miR-28-5p ǂhsa-miR-221 hsa-miR-139-5p ǂhsa-miR-199a-3p hsa-miR-99b ǂhsa-miR-151-5p hsa-miR-766 hsa-miR-142-3p ǂhsa-miR-26a hsa-let-7e ǂhsa-miR-151-3p ǂhsa-miR-30e* ǂhsa-miR-146a ǂhsa-miR-142-5p hsa-miR-1 ǂhsa-miR-152 ǂhsa-miR-22* Up-regulated hsa-miR-1266 hsa-miR-532-5p hsa-miR-29b-1*  P-value  Corrected P-value  0.002  0.048  0.321 0.189 0.260 0.228 0.191 0.174 0.179 0.216 0.259 0.250 0.266 0.389 0.369 0.357 0.344 0.331 0.334 0.366 0.393  0.005 0.015 0.018  0.184 0.271 0.252  0.002 0.004 0.005 0.006 0.007 0.008 0.011 0.016 0.017 0.021 0.032 0.033 0.033 0.034 0.035 0.037 0.042  ǂ Significantly affected by hemolysis  5.2.3 Overlap between surgical and peripheral venous sample sets Comparison of miRNAs significantly differentially expressed in surgical samples and significantly differentially expressed prior to correction in peripheral venous samples identified one miRNA (miR-532-5p) that was up-regulated in both sample sets (Figure 10). The most significantly up-regulated miRNAs (miRNAs with the highest median fold-change) in the 68  surgical samples were miR-375, miR-141, miR-200c, miR-200b, and miR-125b none of which are significantly affected by hemolysis (Table 12). In the set of peripheral venous samples the miR-200 family of miRNAs were present at very low levels or not detected in the majority of samples with miR-200c, miR-141, and miR-200b only detected in 50, 53, and 26 of the 69 samples respectively. While miR-125b and miR-375 were present at higher levels in these samples and were detected in 69 and 68 samples respectively. Of the samples in which these 5 miRNAs were detected, the average normalised Ct was slightly higher in cancer compared to control cases although not significant (Table 14).  69  Figure 10. (A) miR-532 is up-regulated in pulmonary venous (black) samples compared to matched systemic arterial (light grey) samples. (B) miR-532-5p is also slightly up-regulated in lung adenocarcinoma peripheral venous samples compared to controls. Corrected P = 0.27, Mann-Whitney. 70  Table 14. Average normalized Cts from peripheral venous samples of the miRNAs most significantly up-regulated in surgical samples.  hsa-miR-125b  Average Normalized Ct Cancer (Standard Deviation) 0.73 (0.54)  Average Normalized Ct Control (Standard Deviation) 0.54 (0.44)  hsa-miR-375  -1.29 (1.29)  -1.47 (1.17)  hsa-miR-141  -2.71 (1.24)  -2.91 (0.84)  hsa-miR-200c  -2.87 (1.17)  -3.22 (0.75)  hsa-miR-200b  -3.49 (0.75)  -3.61 (0.92)  miRNA  5.3 Discussion A number of studies have been published identifying serum miRNAs that can distinguish NSCLC patients from healthy controls with high sensitivity and specificity (Table 1). However, there is little overlap in the candidate miRNA biomarkers identified in each study. Here, we have used a novel approach to identify tumor-associated miRNAs while also eliminating differential miRNA expression caused by inter-patient variables by comparing two samples collected from the same patient during surgery. In these surgical samples, 35 miRNAs were significantly upregulated and no miRNAs were significantly down-regulated in tumor associated samples. These results are consistent with studies examining miRNA profiles of NSCLC tissue and serum samples which have found a general up-regulation of miRNAs in tumor compared to matched normal samples86, 154. The most significantly up-regulated miRNAs were miR-375, miR-141, miR-200c, miR-200b, and miR-125b (Table 12). Our results support the findings of other studies demonstrating the up-regulation of miR-125b in serum of NSCLC patients93, 98. Studies have suggested both tumor suppressive and oncogenic roles for miR-125b155, 156. The exact role of this 71  miRNA in lung cancer is unclear. However, it has been demonstrated that miR-125b induces apoptosis and plays a tumor suppressive role in lung cancer cell lines157. miR-375 and miR-200b have been shown by other groups to be up-regulated in tissue and sputum but not plasma samples from patients with NSCLC158, 159. MicroRNAs miR-200b, miR-200c, and miR-141 are all members of the miR-200 family (along with miR-200a and mir-429). Interestingly, these three miRNAs were up-regulated in the tumor associated surgical samples compared to the systemic samples and miRs 200a and 429 were detected in 6 tumor associated samples but not detected in matched systemic samples. This family of miRNAs has been shown to inhibit lung adenocarcinoma metastasis160 and have been shown to be over-expressed by ~11-fold in primary lung tumor cell lines compared to cell lines from metastatic sites161. Among the most significantly differentially expressed miRNAs (those with the lowest uncorrected P-values) in the cancer/control 19 miRNAs were up-regulated in the control samples compared to the cancer samples. This finding seems contradictory to the suggestion that lung tumors are releasing miRNAs into circulation. However, we have found that 11 of the 19 miRNAs are significantly up-regulated in samples exhibiting hemolysis (Table 13) suggesting that the control samples used in this study contained more blood cell contaminant miRNAs than the cancer samples. Additionally, 2 of the 3 miRNAs up-regulated in the cancer samples were found to be unaffected by hemolysis supporting this hypothesis. Of note, miR-532-5p was upregulated in cancer samples and was also found to be significantly up-regulated in samples exhibiting hemolysis. If the higher expression of miRNAs in control samples is due to the presence of blood cell miRNAs, this result may be explained by suggesting that the amount of up-regulation in miR-532-5p due to the presence of a tumor is greater than that due to the presence of blood cell contaminants which is supported by the fact that miR-532-5p was also up72  regulated in the surgical samples. Because the most deregulated miRNAs in the surgical samples were shown to be unaffected by hemolysis, it is likely that the amount of hemolysis in these matched samples were relatively equal or that the influence of the tumor on serum miRNA levels is greater than the influence of blood cell miRNAs. The miRNAs most deregulated in the surgical samples showed a slight but not significant increase in expression in cancer versus control samples. The presence of blood-cell associated miRNAs (and possibly miRNAs from other sources) could be interfering with the diseasespecific miRNA profile in these samples. This issue could possibly be solved by examining the miRNA profiles of tumor-derived exosomes isolated from the serum of cancer patients and comparing the results to exosomes isolated from control patients162. Ideally, the miRNA profiles of tumor-derived exosomes should be unaffected by blood cell contaminants. This study uses a novel method to identify serum miRNAs that are up-regulated in the circulation of patients with lung adenocarcinoma however; further research is required to assess the utility of these miRNAs as non-invasive biomarkers.  6. Review of aims Aim 1 of this thesis was to identify candidate circulating miRNA biomarkers for patients with high-risk oral lesions. The results of this study significantly contribute to this field of research as it was the first study to examine global miRNA expression for head and neck malignancy and one of the first attempts to identify miRNA expression in any epithelial precancer. The finding of stage-specific serum miRNAs associated with precancerous lesions provides evidence that circulating miRNA profiles can change with disease progression. These 73  results suggest utility for such miRNAs not only in the early detection of oral cancer but possibly as markers of progression. Additionally, these results imply that serum miRNA signatures derived from patients with invasive cancer in direct contact with the blood stream may not have the same utility in the detection of early stage cancers which are separated by the basement membrane from direct access to the blood vascular system. The identification of miRNAs that are both de-regulated in cancer versus control serum samples and that correspondingly decreases or increases with surgical resection of the tumor support the utility of serum miRNAs as biomarkers for oral cancer detection and disease monitoring. In aim 2 we sought to examine some of the pre-profiling factors that could influence serum miRNA variability as miRNAs significantly affected by these factors are likely to have little utility as biomarkers or endogenous controls for serum miRNA studies. We found that no miRNAs were significantly (P-value < 0.05) differentially expressed in healthy individuals with varying smoking and fasting statuses. However, further studies on are required to confirm these results. We have also found that mechanical hemolysis of blood samples prior to miRNA profiling significantly increases the levels of 130 miRNAs and should be taken into consideration in future circulating miRNA studies. Importantly, we also found that serum miRNA profiles of healthy individuals are stable and show little variation over a range of time periods with the exception of 9 less abundant miRNAs. The results of this study considerably contribute to the field of serum miRNA biomarkers as they highlight the potential and the limitations of serum miRNA biomarkers and call into question some miRNAs that are currently used as endogenous controls and candidate biomarkers in serum miRNA studies. Finally, in aim 3 of this thesis we used a novel approach that avoids inter-patient variability by using the patient as his/her own control to attempt to identify candidate serum 74  miRNAs with utility as lung adenocarcinoma biomarkers. We found 35 miRNAs that were significantly up-regulated in tumor-associated serum samples and 66 miRNAs that were differentially detected in matched tumor-associated and systemic samples. The most highly upregulated miRNAs in the tumor-associated samples have been implicated in NSCLC suggesting the serum miRNA levels in these samples are reflective of tumor miRNA levels. Unfortunately, the most highly up-regulated miRNAs in the tumor-associated surgical samples were not significantly differentially expressed in the cancer/control peripheral venous samples. These results demonstrate the need for alternative methods to isolate miRNAs that are tumor-specific in order to avoid over-whelming the miRNA signatures of patients and controls with miRNAs unrelated to disease-state such as those from blood cell contaminants.  7. Conclusion  7.1 Future perspectives Although we have identified potential biomarkers for the detection of oral and lung cancer, further studies need to be conducted before serum miRNA biomarkers can be implemented in the clinic or used for population screening. For oral cancer biomarkers, future experiments can be designed using previous work in the field of non-invasive cancer biomarker discovery as a guide to identify the analytical and pre-analytical hurdles that need to be overcome. For example, the use of prostate-specific antigen for prostate cancer screening has been problematic due to the fact that this molecule is also elevated in patients with benign prostate conditions7. Therefore, after potential serum miRNA biomarkers are examined in 75  healthy individuals they should also be tested and compared in screening subjects with a range of conditions rather than just those with the disease state of interest. In order to avoid this issue and the issue of data over-fitting, the candidate oral cancer miRNA biomarkers outlined in this thesis will have to be tested in larger cohorts. These cohorts should continue to include individuals who may benefit from an oral cancer screening test (ie. high-risk smokers) and include patients with benign conditions. A more recent study examining the levels of 1205 human miRNAs in serum samples from oesophageal SCC patients identified miR-1246 as the most significantly upregulated miRNA in the serum of cancer patients. Interestingly, the group found that miR-1246 was not differentially expressed in oesophageal SCC tissue samples but was up-regulated in the proximal lymph-nodes of these patients. Additionally, the levels of miR-1266 were higher in isolated exosomes from an oesophageal SCC cell line than in the parent cells suggesting this miRNA is selectively secreted by oesophageal SCC cells and enters the circulatory and lymphatic systems163. Using this study as an example, future work on the candidate miRNAs identified here will include functional studies measuring the levels of these miRNAs in exosomes isolated from OSCC cell lines and possibly serum of patients with OSCC. Additionally, because we have shown that some candidate miRNAs are significantly affected by hemolysis, this issue will have to be addressed before moving on to the validation stage of our oral cancer study. Of the pre-profiling variables examined in this thesis, lysis of blood cells prior to sample processing had the most significant impact on serum miRNA levels. Although an ideal biomarker would be unaffected by such conditions, it is reasonable to consider that one or more of the 130 miRNAs identified in chapter 4 as significantly affected by hemolysis could also exhibit deregulated serum expression due to the presence of a tumor. Therefore, an appealing 76  alternative to excluding these miRNAs from biomarker studies is to establish a method to adjust the quantification of circulating miRNAs according to the amount of hemolysis in the sample. Our results show a significant linear correlation between serum hemoglobin concentration and miRNA levels. Therefore, it is feasible that a correction factor for the quantification of miRNAs may be created using the hemoglobin concentration of the given sample. As part of our study we attempted to create such a correction factor using the 10 matched pairs of hemolysed/nonhemolysed samples from chapter 4. However the coefficients established were not applicable to other data sets. This lack of applicability may be due to the small sample size of this study but also due to the reliability of hemoglobin concentration measurements. Although hemoglobin can be measured spectrophotometrically in serum samples, interference from lipids and other proteins in the sample can lead to less accurate measurements120. Although the spectrophotometer data from the matched lysed and unlysed samples showed little interference from lipids and proteins, the absorbance graphs from the cancer and control samples were irregular demonstrating interference from other compounds. Therefore we were unable to determine a robust correction factor using the available samples. Future studies on blood cell miRNAs using larger sample sizes and more accurate measurements of blood cell lysis, such as a hematology analyzer, may offer a solution to this issue. Unfortunately, the lung adenocarcinoma-associated miRNAs identified in the unique surgical samples used in our study were not significantly differentially expressed in the peripheral venous samples of patients with lung adenocarcinoma compared to samples from healthy controls. One possible reason for these results is that miRNAs released from the tumor are present at very high levels in pulmonary venous effluent directly draining the tumor vasculature bed but are so diluted in systemic circulation that their signals are overwhelmed by 77  other miRNAs such as those from the currently unknown source of circulating miRNAs in healthy individuals. Therefore, new methods of miRNA quantification need to be applied in future studies examining the utility of the miRNAs that are up-regulated in the pulmonary venous samples as biomarkers for lung adenocarcinoma. This analytic noise could be filtered out by isolating and comparing the miRNA profiles of exosomes from serum samples of patients with early stage lung adenocarcinoma to exosomes derived from the serum of healthy controls. Furthermore, limiting miRNA profiling to tumor-derived exosomes expressing certain surface proteins could further hone the list of miRNAs to tumor specific candidates. Alternatively direct miRNA profiling in these samples using more sensitive techniques could also result in more stratified miRNA profiles. One technique that we are currently examining is the use of digital PCR (dPCR) which is more sensitive than qRT-PCR and can directly calculate absolute miRNA concentrations. Our collaborator (Roza Bidshahri) was able to quantify serum miRNAs with a limit of detection of 0.2-0.1 copies/μl (data not shown). This method may allow us to measure differences in miRNA levels of candidate miRNAs that were up-regulated in pulmonary venous samples but were less abundant peripheral samples such as the miR-200 family of miRNAs. Other, more sensitive methods such as next generation sequencing could also be explored102. The results presented in chapter 5 show a modest deregulation of miRNAs in the serum of patients with lung adenocarcinoma. Therefore, future studies will also include the integration of serum miRNA signatures with other screening techniques including low dose spiral computed tomography, risk modeling (smoking pack-years, COPD status, etc.), and autofluorescence bronchoscopy rather than using serum miRNAs as stand-alone biomarkers.  78  7.2 Conclusion The progression of cancer from a localized tumor to a metastatic disease is a devastating event that reduces effective treatment options for patients. Therefore, early detection of cancer is vital to improve the prognosis of patients diagnosed with this disease. Unfortunately, for many cancers, the disease progresses into a late, metastatic stage before symptoms arise and thus noninvasive screening tools are required to ensure the detection of early stage localized tumors. To this end, a number of studies have been conducted examining the utility of circulating miRNAs as biomarkers of disease. Here, we identify serum miRNAs with potential utility as biomarkers for the early detection of oral and lung cancer. Additionally, we’ve examined pre-profiling variables that should be taken into account when conducting serum miRNA biomarker studies. Much more work into standardizing the pre-profiling and analytical aspects of circulating miRNAs as well as functional studies determining their source and function are required before the full clinical potential of circulating miRNAs can be realized. However, the results presented in this thesis provide promising evidence for the utility of circulating miRNAs to improve cancer screening programs.  79  References 1. WHO. Cancer Fact sheet no. 297, 2012. 2. Smith RA, Brooks D, Cokkinides V, Saslow D, Brawley OW. Cancer screening in the United States, 2013: a review of current American Cancer Society guidelines, current issues in cancer screening, and new guidance on cervical cancer screening and lung cancer screening. CA: a cancer journal for clinicians 2013;63:88-105. 3. Chatterjee SK, Zetter BR. Cancer biomarkers: knowing the present and predicting the future. Future Oncol 2005;1:37-50. 4. Rhea JM, Molinaro RJ. Cancer biomarkers: surviving the journey from bench to bedside. MLO: medical laboratory observer 2011;43:10-2, 6, 8; quiz 20, 2. 5. Hanash SM, Baik CS, Kallioniemi O. Emerging molecular biomarkers--blood-based strategies to detect and monitor cancer. Nature reviews. Clinical oncology 2011;8:142-50. 6. Ludwig JA, Weinstein JN. Biomarkers in cancer staging, prognosis and treatment selection. Nature reviews. Cancer 2005;5:845-56. 7. Diamandis EP. Cancer biomarkers: can we turn recent failures into success? Journal of the National Cancer Institute 2010;102:1462-7. 8. Brooks JD. Translational genomics: the challenge of developing cancer biomarkers. Genome research 2012;22:183-7. 9. Warnakulasuriya S. Global epidemiology of oral and oropharyngeal cancer. Oral oncology 2009;45:309-16. 10. Tiziani S, Lopes V, Gunther UL. Early stage diagnosis of oral cancer using 1H NMR-based metabolomics. Neoplasia 2009;11:269-76, 4p following 11. Warnakulasuriya KA, Robinson D, Evans H. Multiple primary tumours following head and neck cancer in southern England during 1961-98. Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology 2003;32:443-9. 12. Mashberg A. Diagnosis of early oral and oropharyngeal squamous carcinoma: obstacles and their amelioration. Oral oncology 2000;36:253-5. 13. Mignogna MD, Fedele S, Lo Russo L. The World Cancer Report and the burden of oral cancer. Eur J Cancer Prev 2004;13:139-42. 14. Fedele S. Diagnostic aids in the screening of oral cancer. Head & neck oncology 2009;1:5. 15. Kujan O, Glenny AM, Duxbury J, Thakker N, Sloan P. Evaluation of screening strategies for improving oral cancer mortality: a Cochrane systematic review. Journal of dental education 2005;69:25565. 16. Sankaranarayanan R, Ramadas K, Thomas G, Muwonge R, Thara S, Mathew B, Rajan B. Effect of screening on oral cancer mortality in Kerala, India: a cluster-randomised controlled trial. Lancet 2005;365:1927-33. 17. Scheifele C, Schmidt-Westhausen AM, Dietrich T, Reichart PA. The sensitivity and specificity of the OralCDx technique: evaluation of 103 cases. Oral oncology 2004;40:824-8. 18. Lingen MW, Kalmar JR, Karrison T, Speight PM. Critical evaluation of diagnostic aids for the detection of oral cancer. Oral oncology 2008;44:10-22. 19. Poh CF, MacAulay CE, Zhang L, Rosin MP. Tracing the "at-risk" oral mucosa field with autofluorescence: steps toward clinical impact. Cancer Prev Res (Phila) 2009;2:401-4. 20. Rethman MP, Carpenter W, Cohen EE, Epstein J, Evans CA, Flaitz CM, Graham FJ, Hujoel PP, Kalmar JR, Koch WM, Lambert PM, Lingen MW, et al. Evidence-based clinical recommendations regarding screening for oral squamous cell carcinomas. J Am Dent Assoc 2010;141:509-20. 21. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA: a cancer journal for clinicians 2011;61:69-90. 80  22. Goldstraw P, Ball D, Jett JR, Le Chevalier T, Lim E, Nicholson AG, Shepherd FA. Nonsmall-cell lung cancer. Lancet 2011;378:1727-40. 23. Crino L, Weder W, van Meerbeeck J, Felip E. Early stage and locally advanced (nonmetastatic) non-small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO 2010;21 Suppl 5:v103-15. 24. Wood DE, Eapen GA, Ettinger DS, Hou L, Jackman D, Kazerooni E, Klippenstein D, Lackner RP, Leard L, Leung AN, Massion PP, Meyers BF, et al. Lung cancer screening. Journal of the National Comprehensive Cancer Network : JNCCN 2012;10:240-65. 25. Melamed MR, Flehinger BJ, Zaman MB, Heelan RT, Perchick WA, Martini N. Screening for early lung cancer. Results of the Memorial Sloan-Kettering study in New York. Chest 1984;86:44-53. 26. Kubik A, Parkin DM, Khlat M, Erban J, Polak J, Adamec M. Lack of benefit from semiannual screening for cancer of the lung: follow-up report of a randomized controlled trial on a population of high-risk males in Czechoslovakia. International journal of cancer. Journal international du cancer 1990;45:26-33. 27. Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, Fagerstrom RM, Gareen IF, Gatsonis C, Marcus PM, Sicks JD. Reduced lung-cancer mortality with low-dose computed tomographic screening. The New England journal of medicine 2011;365:395-409. 28. Van't Westeinde SC, van Klaveren RJ. Screening and early detection of lung cancer. Cancer J 2011;17:3-10. 29. Haussinger K, Becker H, Stanzel F, Kreuzer A, Schmidt B, Strausz J, Cavaliere S, Herth F, Kohlhaufl M, Muller KM, Huber RM, Pichlmeier U, et al. Autofluorescence bronchoscopy with white light bronchoscopy compared with white light bronchoscopy alone for the detection of precancerous lesions: a European randomised controlled multicentre trial. Thorax 2005;60:496-503. 30. Pawluk HC, Short MA, Lam S, McWilliams AM, Ionescue DN, Zeng H: Improvements to a laser Raman spectroscopy system for reducing the false positives of autofluorescence bronchoscopies Proc. of SPIE Vol 2012; 8207:820737-1. 31. Li G, Guillaud M, LeRiche J, McWilliams A, Gazdar A, Lam S, MacAulay C. Automated sputum cytometry for detection of intraepithelial neoplasias in the lung. Anal Cell Pathol (Amst) 2012;35:187-201. 32. Ostrow KL, Hoque MO, Loyo M, Brait M, Greenberg A, Siegfried JM, Grandis JR, Gaither Davis A, Bigbee WL, Rom W, Sidransky D. Molecular analysis of plasma DNA for the early detection of lung cancer by quantitative methylation-specific PCR. Clinical cancer research : an official journal of the American Association for Cancer Research 2010;16:3463-72. 33. Xiao T, Ying W, Li L, Hu Z, Ma Y, Jiao L, Ma J, Cai Y, Lin D, Guo S, Han N, Di X, et al. An approach to studying lung cancer-related proteins in human blood. Molecular & cellular proteomics : MCP 2005;4:1480-6. 34. Carpagnano GE, Foschino-Barbaro MP, Mule G, Resta O, Tommasi S, Mangia A, Carpagnano F, Stea G, Susca A, Di Gioia G, De Lena M, Paradiso A. 3p microsatellite alterations in exhaled breath condensate from patients with non-small cell lung cancer. American journal of respiratory and critical care medicine 2005;172:738-44. 35. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004;116:28197. 36. Lee Y, Ahn C, Han J, Choi H, Kim J, Yim J, Lee J, Provost P, Radmark O, Kim S, Kim VN. The nuclear RNase III Drosha initiates microRNA processing. Nature 2003;425:415-9. 37. Lund E, Guttinger S, Calado A, Dahlberg JE, Kutay U. Nuclear export of microRNA precursors. Science 2004;303:95-8. 38. Ameres SL, Martinez J, Schroeder R. Molecular basis for target RNA recognition and cleavage by human RISC. Cell 2007;130:101-12. 81  39. Yang JS, Phillips MD, Betel D, Mu P, Ventura A, Siepel AC, Chen KC, Lai EC. Widespread regulatory activity of vertebrate microRNA* species. RNA 2011;17:312-26. 40. Filipowicz W, Bhattacharyya SN, Sonenberg N. Mechanisms of post-transcriptional regulation by microRNAs: are the answers in sight? Nature reviews. Genetics 2008;9:102-14. 41. Pillai RS, Bhattacharyya SN, Filipowicz W. Repression of protein synthesis by miRNAs: how many mechanisms? Trends in cell biology 2007;17:118-26. 42. Huntzinger E, Izaurralde E. Gene silencing by microRNAs: contributions of translational repression and mRNA decay. Nature reviews. Genetics 2011;12:99-110. 43. Lim LP, Lau NC, Garrett-Engele P, Grimson A, Schelter JM, Castle J, Bartel DP, Linsley PS, Johnson JM. Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 2005;433:769-73. 44. Piao X, Zhang X, Wu L, Belasco JG. CCR4-NOT deadenylates mRNA associated with RNAinduced silencing complexes in human cells. Molecular and cellular biology 2010;30:1486-94. 45. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 2005;120:15-20. 46. Brennecke J, Cohen SM. Towards a complete description of the microRNA complement of animal genomes. Genome Biol 2003;4:228. 47. Chen CZ, Li L, Lodish HF, Bartel DP. MicroRNAs modulate hematopoietic lineage differentiation. Science 2004;303:83-6. 48. Dostie J, Mourelatos Z, Yang M, Sharma A, Dreyfuss G. Numerous microRNPs in neuronal cells containing novel microRNAs. Rna 2003;9:180-6. 49. Ioshikhes I, Roy S, Sen CK. Algorithms for mapping of mRNA targets for microRNA. DNA Cell Biol 2007;26:265-72. 50. Reinhart BJ, Slack FJ, Basson M, Pasquinelli AE, Bettinger JC, Rougvie AE, Horvitz HR, Ruvkun G. The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature 2000;403:901-6. 51. Xu P, Vernooy SY, Guo M, Hay BA. The Drosophila microRNA Mir-14 suppresses cell death and is required for normal fat metabolism. Curr Biol 2003;13:790-5. 52. Iorio MV, Croce CM. MicroRNA dysregulation in cancer: diagnostics, monitoring and therapeutics. A comprehensive review. EMBO molecular medicine 2012;4:143-59. 53. Calin GA, Sevignani C, Dumitru CD, Hyslop T, Noch E, Yendamuri S, Shimizu M, Rattan S, Bullrich F, Negrini M, Croce CM. Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers. Proceedings of the National Academy of Sciences of the United States of America 2004;101:2999-3004. 54. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646-74. 55. Negrini M, Nicoloso MS, Calin GA. MicroRNAs and cancer--new paradigms in molecular oncology. Current opinion in cell biology 2009;21:470-9. 56. Lu J, Getz G, Miska EA, Alvarez-Saavedra E, Lamb J, Peck D, Sweet-Cordero A, Ebert BL, Mak RH, Ferrando AA, Downing JR, Jacks T, et al. MicroRNA expression profiles classify human cancers. Nature 2005;435:834-8. 57. Nana-Sinkam SP, Croce CM. Clinical Applications for microRNAs in Cancer. Clinical pharmacology and therapeutics 2013;93:98-104. 58. Calin GA, Cimmino A, Fabbri M, Ferracin M, Wojcik SE, Shimizu M, Taccioli C, Zanesi N, Garzon R, Aqeilan RI, Alder H, Volinia S, et al. MiR-15a and miR-16-1 cluster functions in human leukemia. Proceedings of the National Academy of Sciences of the United States of America 2008;105:5166-71. 59. Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY. miR-21-mediated tumor growth. Oncogene 2007;26:2799-803. 60. Liu Q, Fu H, Xing R, Tie Y, Zhu J, Sun Z, Zheng X. Survivin knockdown combined with apoptin overexpression inhibits cell growth significantly. Cancer biology & therapy 2008;7:1053-60. 82  61. Gandellini P, Profumo V, Folini M, Zaffaroni N. MicroRNAs as new therapeutic targets and tools in cancer. Expert opinion on therapeutic targets 2011;15:265-79. 62. Hsu SD, Chu CH, Tsou AP, Chen SJ, Chen HC, Hsu PW, Wong YH, Chen YH, Chen GH, Huang HD. miRNAMap 2.0: genomic maps of microRNAs in metazoan genomes. Nucleic acids research 2008;36:D165-9. 63. Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogosova-Agadjanyan EL, Peterson A, Noteboom J, O'Briant KC, Allen A, Lin DW, Urban N, et al. Circulating microRNAs as stable blood-based markers for cancer detection. Proceedings of the National Academy of Sciences of the United States of America 2008;105:10513-8. 64. Gallo A, Tandon M, Alevizos I, Illei GG. The Majority of MicroRNAs Detectable in Serum and Saliva Is Concentrated in Exosomes. PloS one 2012;7:e30679. 65. Diehl P, Fricke A, Sander L, Stamm J, Bassler N, Htun N, Ziemann M, Helbing T, El-Osta A, Jowett JB, Peter K. Microparticles: major transport vehicles for distinct microRNAs in circulation. Cardiovascular research 2012;93:633-44. 66. Vickers KC, Palmisano BT, Shoucri BM, Shamburek RD, Remaley AT. MicroRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins. Nature cell biology 2011;13:423-33. 67. Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson DF, Mitchell PS, Bennett CF, Pogosova-Agadjanyan EL, Stirewalt DL, Tait JF, Tewari M. Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma. Proceedings of the National Academy of Sciences of the United States of America 2011;108:5003-8. 68. Thery C, Zitvogel L, Amigorena S. Exosomes: composition, biogenesis and function. Nature reviews. Immunology 2002;2:569-79. 69. Chiba M, Kimura M, Asari S. Exosomes secreted from human colorectal cancer cell lines contain mRNAs, microRNAs and natural antisense RNAs, that can transfer into the human hepatoma HepG2 and lung cancer A549 cell lines. Oncology reports 2012;28:1551-8. 70. Yang M, Chen J, Su F, Yu B, Lin L, Liu Y, Huang JD, Song E. Microvesicles secreted by macrophages shuttle invasion-potentiating microRNAs into breast cancer cells. Molecular cancer 2011;10:117. 71. Hugel B, Martinez MC, Kunzelmann C, Freyssinet JM. Membrane microparticles: two sides of the coin. Physiology (Bethesda) 2005;20:22-7. 72. Zhang L, Song J, Cavigiolio G, Ishida BY, Zhang S, Kane JP, Weisgraber KH, Oda MN, Rye KA, Pownall HJ, Ren G. Morphology and structure of lipoproteins revealed by an optimized negativestaining protocol of electron microscopy. Journal of lipid research 2011;52:175-84. 73. Turchinovich A, Weiz L, Burwinkel B. Extracellular miRNAs: the mystery of their origin and function. Trends in biochemical sciences 2012;37:460-5. 74. Creemers EE, Tijsen AJ, Pinto YM. Circulating microRNAs: novel biomarkers and extracellular communicators in cardiovascular disease? Circulation research 2012;110:483-95. 75. Wang JF, Yu ML, Yu G, Bian JJ, Deng XM, Wan XJ, Zhu KM. Serum miR-146a and miR223 as potential new biomarkers for sepsis. Biochemical and biophysical research communications 2010;394:184-8. 76. Wang K, Zhang S, Marzolf B, Troisch P, Brightman A, Hu Z, Hood LE, Galas DJ. Circulating microRNAs, potential biomarkers for drug-induced liver injury. Proceedings of the National Academy of Sciences of the United States of America 2009;106:4402-7. 77. Zampetaki A, Kiechl S, Drozdov I, Willeit P, Mayr U, Prokopi M, Mayr A, Weger S, Oberhollenzer F, Bonora E, Shah A, Willeit J, et al. Plasma microRNA profiling reveals loss of endothelial miR-126 and other microRNAs in type 2 diabetes. Circulation research 2010;107:810-7. 78. Kosaka N, Iguchi H, Ochiya T. Circulating microRNA in body fluid: a new potential biomarker for cancer diagnosis and prognosis. Cancer science 2010;101:2087-92. 83  79. Wong TS, Liu XB, Wong BY, Ng RW, Yuen AP, Wei WI. Mature miR-184 as Potential Oncogenic microRNA of Squamous Cell Carcinoma of Tongue. Clinical cancer research : an official journal of the American Association for Cancer Research 2008;14:2588-92. 80. Lin SC, Liu CJ, Lin JA, Chiang WF, Hung PS, Chang KW. miR-24 up-regulation in oral carcinoma: positive association from clinical and in vitro analysis. Oral oncology 2010;46:204-8. 81. Komatsu S, Ichikawa D, Takeshita H, Tsujiura M, Morimura R, Nagata H, Kosuga T, Iitaka D, Konishi H, Shiozaki A, Fujiwara H, Okamoto K, et al. Circulating microRNAs in plasma of patients with oesophageal squamous cell carcinoma. British journal of cancer 2011;105:104-11. 82. Hsu CM, Lin PM, Wang YM, Chen ZJ, Lin SF, Yang MY. Circulating miRNA is a novel marker for head and neck squamous cell carcinoma. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 2012;33:1933-42. 83. Yang CC, Hung PS, Wang PW, Liu CJ, Chu TH, Cheng HW, Lin SC. miR-181 as a putative biomarker for lymph-node metastasis of oral squamous cell carcinoma. Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology 2011;40:397-404. 84. Liu CJ, Kao SY, Tu HF, Tsai MM, Chang KW, Lin SC. Increase of microRNA miR-31 level in plasma could be a potential marker of oral cancer. Oral diseases 2010;16:360-4. 85. Lu YC, Chen YJ, Wang HM, Tsai CY, Chen WH, Huang YC, Fan KH, Tsai CN, Huang SF, Kang CJ, Chang JT, Cheng AJ. Oncogenic function and early detection potential of miRNA-10b in oral cancer as identified by microRNA profiling. Cancer Prev Res (Phila) 2012;5:665-74. 86. Chen X, Hu Z, Wang W, Ba Y, Ma L, Zhang C, Wang C, Ren Z, Zhao Y, Wu S, Zhuang R, Zhang Y, et al. Identification of ten serum microRNAs from a genome-wide serum microRNA expression profile as novel noninvasive biomarkers for nonsmall cell lung cancer diagnosis. International journal of cancer. Journal international du cancer 2012;130:1620-8. 87. Boeri M, Verri C, Conte D, Roz L, Modena P, Facchinetti F, Calabro E, Croce CM, Pastorino U, Sozzi G. MicroRNA signatures in tissues and plasma predict development and prognosis of computed tomography detected lung cancer. Proceedings of the National Academy of Sciences of the United States of America 2011;108:3713-8. 88. Bianchi F, Nicassio F, Marzi M, Belloni E, Dall'olio V, Bernard L, Pelosi G, Maisonneuve P, Veronesi G, Di Fiore PP. A serum circulating miRNA diagnostic test to identify asymptomatic high-risk individuals with early stage lung cancer. EMBO molecular medicine 2011;3:495-503. 89. Foss KM, Sima C, Ugolini D, Neri M, Allen KE, Weiss GJ. miR-1254 and miR-574-5p: serum-based microRNA biomarkers for early-stage non-small cell lung cancer. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer 2011;6:4828. 90. Liu XG, Zhu WY, Huang YY, Ma LN, Zhou SQ, Wang YK, Zeng F, Zhou JH, Zhang YK. High expression of serum miR-21 and tumor miR-200c associated with poor prognosis in patients with lung cancer. Med Oncol 2012;29:618-26. 91. Wang ZX, Bian HB, Wang JR, Cheng ZX, Wang KM, De W. Prognostic significance of serum miRNA-21 expression in human non-small cell lung cancer. Journal of surgical oncology 2011;104:847-51. 92. Keller A, Leidinger P, Gislefoss R, Haugen A, Langseth H, Staehler P, Lenhof HP, Meese E. Stable serum miRNA profiles as potential tool for non-invasive lung cancer diagnosis. RNA biology 2011;8:506-16. 93. Yuxia M, Zhennan T, Wei Z. Circulating miR-125b is a novel biomarker for screening nonsmall-cell lung cancer and predicts poor prognosis. Journal of cancer research and clinical oncology 2012;138:2045-50. 94. Hennessey PT, Sanford T, Choudhary A, Mydlarz WW, Brown D, Adai AT, Ochs MF, Ahrendt SA, Mambo E, Califano JA. Serum microRNA biomarkers for detection of non-small cell lung cancer. PloS one 2012;7:e32307. 84  95. Heegaard NH, Schetter AJ, Welsh JA, Yoneda M, Bowman ED, Harris CC. Circulating micro-RNA expression profiles in early stage nonsmall cell lung cancer. International journal of cancer. Journal international du cancer 2012;130:1378-86. 96. Le HB, Zhu WY, Chen DD, He JY, Huang YY, Liu XG, Zhang YK. Evaluation of dynamic change of serum miR-21 and miR-24 in pre- and post-operative lung carcinoma patients. Med Oncol 2012;29:3190-7. 97. Roth C, Stuckrath I, Pantel K, Izbicki JR, Tachezy M, Schwarzenbach H. Low levels of cellfree circulating miR-361-3p and miR-625* as blood-based markers for discriminating malignant from benign lung tumors. PloS one 2012;7:e38248. 98. Cui EH, Li HJ, Hua F, Wang B, Mao W, Feng XR, Li JY, Wang X. Serum microRNA 125b as a diagnostic or prognostic biomarker for advanced NSCLC patients receiving cisplatin-based chemotherapy. Acta pharmacologica Sinica 2013;34:309-13. 99. Lin Q, Mao W, Shu Y, Lin F, Liu S, Shen H, Gao W, Li S, Shen D. A cluster of specified microRNAs in peripheral blood as biomarkers for metastatic non-small-cell lung cancer by stem-loop RTPCR. Journal of cancer research and clinical oncology 2012;138:85-93. 100. Munker R, Calin GA. MicroRNA profiling in cancer. Clin Sci (Lond) 2011;121:141-58. 101. Zampetaki A, Mayr M. Analytical challenges and technical limitations in assessing circulating miRNAs. Thrombosis and haemostasis 2012;108:592-8. 102. Schweiger MR, Kerick M, Timmermann B, Isau M. The power of NGS technologies to delineate the genome organization in cancer: from mutations to structural variations and epigenetic alterations. Cancer metastasis reviews 2011;30:199-210. 103. Kosaka N, Iguchi H, Ochiya T. Circulating microRNA in body fluid: a new potential biomarker for cancer diagnosis and prognosis. Cancer science 2010;101:2087-92. 104. Kuwabara Y, Ono K, Horie T, Nishi H, Nagao K, Kinoshita M, Watanabe S, Baba O, Kojima Y, Shizuta S. Increased MicroRNA-1 and MicroRNA-133a Levels in Serum of Patients With Cardiovascular Disease Indicate Myocardial DamageClinical Perspective. Circulation: Cardiovascular Genetics 2011;4:446-54. 105. Zhu HT, Dong QZ, Wang G, Zhou HJ, Ren N, Jia HL, Ye QH, Qin LX. Identification of suitable reference genes for qRT-PCR analysis of circulating microRNAs in hepatitis B virus-infected patients. Molecular biotechnology 2012;50:49-56. 106. Davoren PA, McNeill RE, Lowery AJ, Kerin MJ, Miller N. Identification of suitable endogenous control genes for microRNA gene expression analysis in human breast cancer. BMC molecular biology 2008;9:76. 107. Shen J, Todd NW, Zhang H, Yu L, Lingxiao X, Mei Y, Guarnera M, Liao J, Chou A, Lu CL, Jiang Z, Fang H, et al. Plasma microRNAs as potential biomarkers for non-small-cell lung cancer. Laboratory investigation; a journal of technical methods and pathology 2011;91:579-87. 108. Zuo Z, Calin GA, de Paula HM, Medeiros LJ, Fernandez MH, Shimizu M, Garcia-Manero G, Bueso-Ramos CE. Circulating microRNAs let-7a and miR-16 predict progression-free survival and overall survival in patients with myelodysplastic syndrome. Blood 2011;118:413-5. 109. Mahn R, Heukamp LC, Rogenhofer S, von Ruecker A, Muller SC, Ellinger J. Circulating microRNAs (miRNA) in serum of patients with prostate cancer. Urology 2011;77:1265 e9-16. 110. Mestdagh P, Van Vlierberghe P, De Weer A, Muth D, Westermann F, Speleman F, Vandesompele J. A novel and universal method for microRNA RT-qPCR data normalization. Genome biology 2009;10:R64. 111. Heneghan HM, Miller N, Lowery AJ, Sweeney KJ, Newell J, Kerin MJ. Circulating microRNAs as novel minimally invasive biomarkers for breast cancer. Annals of surgery 2010;251:499505. 112. Ng EK, Chong WW, Jin H, Lam EK, Shin VY, Yu J, Poon TC, Ng SS, Sung JJ. Differential expression of microRNAs in plasma of patients with colorectal cancer: a potential marker for colorectal cancer screening. Gut 2009;58:1375-81. 85  113. Zhang Y, Jia Y, Zheng R, Guo Y, Wang Y, Guo H, Fei M, Sun S. Plasma microRNA-122 as a biomarker for viral-, alcohol-, and chemical-related hepatic diseases. Clinical chemistry 2010;56:18308. 114. Zampetaki A, Willeit P, Drozdov I, Kiechl S, Mayr M. Profiling of circulating microRNAs: from single biomarkers to re-wired networks. Cardiovascular research 2012;93:555-62. 115. Pavlou MP, Diamandis EP. The cancer cell secretome: a good source for discovering biomarkers? Journal of proteomics 2010;73:1896-906. 116. Zhang J, Zhao H, Gao Y, Zhang W. Secretory miRNAs as novel cancer biomarkers. Biochimica et biophysica acta 2012;1826:32-43. 117. Ransohoff DF. Bias as a threat to the validity of cancer molecular-marker research. Nature reviews. Cancer 2005;5:142-9. 118. Yee J, Sadar MD, Sin DD, Kuzyk M, Xing L, Kondra J, McWilliams A, Man SF, Lam S. Connective tissue-activating peptide III: a novel blood biomarker for early lung cancer detection. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2009;27:2787-92. 119. Dimeski G. Effects of hemolysis on the Roche ammonia method for Hitachi analyzers. Clinical chemistry 2004;50:976-7. 120. Noe DA, Weedn V, Bell WR. Direct spectrophotometry of serum hemoglobin: an Allen correction compared with a three-wavelength polychromatic analysis. Clinical chemistry 1984;30:627-30. 121. Andreasen D, Fog JU, Biggs W, Salomon J, Dahslveen IK, Baker A, Mouritzen P. Improved microRNA quantification in total RNA from clinical samples. Methods 2010;50:S6-9. 122. Chari R, Lonergan KM, Pikor LA, Coe BP, Zhu CQ, Chan TH, MacAulay CE, Tsao MS, Lam S, Ng RT, Lam WL. A sequence-based approach to identify reference genes for gene expression analysis. BMC medical genomics 2010;3:32. 123. Sing T, Sander O, Beerenwinkel N, Lengauer T. ROCR: visualizing classifier performance in R. Bioinformatics 2005;21:3940-1. 124. Ohshima K, Inoue K, Fujiwara A, Hatakeyama K, Kanto K, Watanabe Y, Muramatsu K, Fukuda Y, Ogura S, Yamaguchi K, Mochizuki T. Let-7 microRNA family is selectively secreted into the extracellular environment via exosomes in a metastatic gastric cancer cell line. PloS one 2010;5:e13247. 125. Kosaka N, Iguchi H, Yoshioka Y, Hagiwara K, Takeshita F, Ochiya T. Competitive interactions of cancer cells and normal cells via secretory microRNAs. The Journal of biological chemistry 2012;287:1397-405. 126. McDonald JS, Milosevic D, Reddi HV, Grebe SK, Algeciras-Schimnich A. Analysis of circulating microRNA: preanalytical and analytical challenges. Clinical chemistry 2011;57:833-40. 127. Cervigne NK, Reis PP, Machado J, Sadikovic B, Bradley G, Galloni NN, Pintilie M, Jurisica I, Perez-Ordonez B, Gilbert R, Gullane P, Irish J, et al. Identification of a microRNA signature associated with progression of leukoplakia to oral carcinoma. Human molecular genetics 2009;18:4818-29. 128. Takamizawa J, Konishi H, Yanagisawa K, Tomida S, Osada H, Endoh H, Harano T, Yatabe Y, Nagino M, Nimura Y, Mitsudomi T, Takahashi T. Reduced expression of the let-7 microRNAs in human lung cancers in association with shortened postoperative survival. Cancer research 2004;64:37536. 129. Johnson SM, Grosshans H, Shingara J, Byrom M, Jarvis R, Cheng A, Labourier E, Reinert KL, Brown D, Slack FJ. RAS is regulated by the let-7 microRNA family. Cell 2005;120:635-47. 130. Aqeilan RI, Calin GA, Croce CM. miR-15a and miR-16-1 in cancer: discovery, function and future perspectives. Cell death and differentiation 2010;17:215-20. 131. Lin RJ, Xiao DW, Liao LD, Chen T, Xie ZF, Huang WZ, Wang WS, Jiang TF, Wu BL, Li EM, Xu LY. MiR-142-3p as a potential prognostic biomarker for esophageal squamous cell carcinoma. Journal of surgical oncology 2012;105:175-82. 132. Kozaki K, Imoto I, Mogi S, Omura K, Inazawa J. Exploration of tumor-suppressive microRNAs silenced by DNA hypermethylation in oral cancer. Cancer research 2008;68:2094-105. 86  133. Kikkawa N, Hanazawa T, Fujimura L, Nohata N, Suzuki H, Chazono H, Sakurai D, Horiguchi S, Okamoto Y, Seki N. miR-489 is a tumour-suppressive miRNA target PTPN11 in hypopharyngeal squamous cell carcinoma (HSCC). British journal of cancer 2010;103:877-84. 134. Zhang L, Poh CF, Williams M, Laronde DM, Berean K, Gardner PJ, Jiang H, Wu L, Lee JJ, Rosin MP. Loss of heterozygosity (LOH) profiles--validated risk predictors for progression to oral cancer. Cancer Prev Res (Phila) 2012;5:1081-9. 135. Macaulay C, Poh CF, Guillaud M, Michele Williams P, Laronde DM, Zhang L, Rosin MP. High throughput image cytometry for detection of suspicious lesions in the oral cavity. Journal of biomedical optics 2012;17:086004-1. 136. Hunter MP, Ismail N, Zhang X, Aguda BD, Lee EJ, Yu L, Xiao T, Schafer J, Lee ML, Schmittgen TD, Nana-Sinkam SP, Jarjoura D, et al. Detection of microRNA expression in human peripheral blood microvesicles. PloS one 2008;3:e3694. 137. Duttagupta R, Jiang R, Gollub J, Getts RC, Jones KW. Impact of cellular miRNAs on circulating miRNA biomarker signatures. PloS one 2011;6:e20769. 138. Pritchard CC, Kroh E, Wood B, Arroyo JD, Dougherty KJ, Miyaji MM, Tait JF, Tewari M. Blood cell origin of circulating microRNAs: a cautionary note for cancer biomarker studies. Cancer Prev Res (Phila) 2012;5:492-7. 139. Kroh EM, Parkin RK, Mitchell PS, Tewari M. Analysis of circulating microRNA biomarkers in plasma and serum using quantitative reverse transcription-PCR (qRT-PCR). Methods 2010;50:298-301. 140. Parkin DM, Pisani P, Lopez AD, Masuyer E. At least one in seven cases of cancer is caused by smoking. Global estimates for 1985. International journal of cancer. Journal international du cancer 1994;59:494-504. 141. Huang LX, Wu XH. [Effect of Xinmailong on hypoxia-inducible factor-1alpha expression in neonatal rats with asphyxia]. Zhongguo dang dai er ke za zhi = Chinese journal of contemporary pediatrics 2009;11:683-6. 142. Lippi G, Salvagno GL, Montagnana M, Brocco G, Guidi GC. Influence of hemolysis on routine clinical chemistry testing. Clinical chemistry and laboratory medicine : CCLM / FESCC 2006;44:311-6. 143. Mo MH, Chen L, Fu Y, Wang W, Fu SW. Cell-free Circulating miRNA Biomarkers in Cancer. Journal of Cancer 2012;3:432-48. 144. Bihrer V, Friedrich-Rust M, Kronenberger B, Forestier N, Haupenthal J, Shi Y, PevelingOberhag J, Radeke HH, Sarrazin C, Herrmann E, Zeuzem S, Waidmann O, et al. Serum miR-122 as a biomarker of necroinflammation in patients with chronic hepatitis C virus infection. The American journal of gastroenterology 2011;106:1663-9. 145. Xu J, Wu C, Che X, Wang L, Yu D, Zhang T, Huang L, Li H, Tan W, Wang C, Lin D. Circulating microRNAs, miR-21, miR-122, and miR-223, in patients with hepatocellular carcinoma or chronic hepatitis. Molecular carcinogenesis 2011;50:136-42. 146. Zhang L, Hou D, Chen X, Li D, Zhu L, Zhang Y, Li J, Bian Z, Liang X, Cai X, Yin Y, Wang C, et al. Exogenous plant MIR168a specifically targets mammalian LDLRAP1: evidence of crosskingdom regulation by microRNA. Cell research 2012;22:107-26. 147. Blondal T, Jensby Nielsen S, Baker A, Andreasen D, Mouritzen P, Wrang Teilum M, Dahlsveen IK. Assessing sample and miRNA profile quality in serum and plasma or other biofluids. Methods 2012. 148. Hawkins R. Discrepancy between visual and spectrophotometric assessment of sample haemolysis. Annals of clinical biochemistry 2002;39:521-2. 149. Sharp MK, Mohammad SF. Hemolysis in needleless connectors for phlebotomy. ASAIO J 2003;49:128-30.  87  150. Zhan M, Miller CP, Papayannopoulou T, Stamatoyannopoulos G, Song CZ. MicroRNA expression dynamics during murine and human erythroid differentiation. Experimental hematology 2007;35:1015-25. 151. Witwer KW. XenomiRs and miRNA homeostasis in health and disease: evidence that diet and dietary miRNAs directly and indirectly influence circulating miRNA profiles. RNA biology 2012;9:1147-54. 152. Huang Y, Dai Y, Zhang J, Wang C, Li D, Cheng J, Lu Y, Ma K, Tan L, Xue F, Qin B. Circulating microRNAs as potential biomarkers for smoking-related interstitial fibrosis. Biomarkers : biochemical indicators of exposure, response, and susceptibility to chemicals 2012;17:435-40. 153. Gilad S, Meiri E, Yogev Y, Benjamin S, Lebanony D, Yerushalmi N, Benjamin H, Kushnir M, Cholakh H, Melamed N, Bentwich Z, Hod M, et al. Serum microRNAs are promising novel biomarkers. PloS one 2008;3:e3148. 154. Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M, Prueitt RL, Yanaihara N, et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proceedings of the National Academy of Sciences of the United States of America 2006;103:2257-61. 155. Guan Y, Yao H, Zheng Z, Qiu G, Sun K. MiR-125b targets BCL3 and suppresses ovarian cancer proliferation. International journal of cancer. Journal international du cancer 2011;128:2274-83. 156. Pogue AI, Cui JG, Li YY, Zhao Y, Culicchia F, Lukiw WJ. Micro RNA-125b (miRNA125b) function in astrogliosis and glial cell proliferation. Neuroscience letters 2010;476:18-22. 157. Gong J, Zhang JP, Li B, Zeng C, You K, Chen MX, Yuan Y, Zhuang SM. MicroRNA-125b promotes apoptosis by regulating the expression of Mcl-1, Bcl-w and IL-6R. Oncogene 2012. 158. Yu L, Todd NW, Xing L, Xie Y, Zhang H, Liu Z, Fang H, Zhang J, Katz RL, Jiang F. Early detection of lung adenocarcinoma in sputum by a panel of microRNA markers. International journal of cancer. Journal international du cancer 2010;127:2870-8. 159. Shen J, Liu Z, Todd NW, Zhang H, Liao J, Yu L, Guarnera MA, Li R, Cai L, Zhan M, Jiang F. Diagnosis of lung cancer in individuals with solitary pulmonary nodules by plasma microRNA biomarkers. BMC cancer 2011;11:374. 160. Roybal JD, Zang Y, Ahn YH, Yang Y, Gibbons DL, Baird BN, Alvarez C, Thilaganathan N, Liu DD, Saintigny P, Heymach JV, Creighton CJ, et al. miR-200 Inhibits lung adenocarcinoma cell invasion and metastasis by targeting Flt1/VEGFR1. Molecular cancer research : MCR 2011;9:25-35. 161. Gibbons DL, Lin W, Creighton CJ, Rizvi ZH, Gregory PA, Goodall GJ, Thilaganathan N, Du L, Zhang Y, Pertsemlidis A, Kurie JM. Contextual extracellular cues promote tumor cell EMT and metastasis by regulating miR-200 family expression. Genes & development 2009;23:2140-51. 162. Rabinowits G, Gercel-Taylor C, Day JM, Taylor DD, Kloecker GH. Exosomal microRNA: a diagnostic marker for lung cancer. Clinical lung cancer 2009;10:42-6. 163. Takeshita N, Hoshino I, Mori M, Akutsu Y, Hanari N, Yoneyama Y, Ikeda N, Isozaki Y, Maruyama T, Akanuma N, Komatsu A, Jitsukawa M, et al. Serum microRNA expression profile: miR1246 as a novel diagnostic and prognostic biomarker for oesophageal squamous cell carcinoma. British journal of cancer 2013;108:644-52.  88  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073891/manifest

Comment

Related Items