{"http:\/\/dx.doi.org\/10.14288\/1.0090340":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Dentistry, Faculty of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeCampus":[{"value":"UBCV","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"An, Ding","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2009-08-17T19:45:15Z","type":"literal","lang":"en"},{"value":"2002","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#relatedDegree":[{"value":"Master of Science - MSc","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeGrantor":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"Oral squamous carcinomas (SCC) remain a significant public health problem worldwide\r\ndespite advances in therapy and local disease control. New innovative strategies must be\r\ndeveloped for the prevention, early detection and treatment of oral carcinoma. Such\r\napproaches will be heavily dependent on a better understanding of the molecular\r\nmechanisms underlying carcinogenesis at this site.\r\nThis thesis describes a series of studies done on oral cancers and premalignant lesions to\r\nbetter define the role of alterations on chromosome 11 in the development of oral cancer\r\nat this site. Although numerous studies have reported the presence of alterations on this\r\nchromosome arm in oral cancers, few studies have examined premalignant lesions in\r\norder to determine whether such alterations play a role in the development of the disease.\r\nThe objectives of this thesis were: 1) to use microsatellite analysis to examine DNA\r\nextracted from severe dysplasia, carcinoma in situ (CIS) and SCC for novel alterations on\r\n11q;2) to determine at what stage of oral cancer development the alteration occurred by\r\nperforming microsatellite analysis on a spectrum of stages of oral premalignant lesions\r\n(hyperplasia, mild dysplasia, moderate dysplasia, severe dysplasia, CIS) as well as\r\ninvasive SCC. The data obtained was compared to 2 previously studied hotspots on 11q:\r\nthe int2 (11q13) and D11S1778 (11q22-23); and 3) to determine the significance of\r\nallelic imbalance (Al) at int2 (11q13 ) and D11S1778 (11q22-23) to the progression of\r\n\r\noral premalignant lesions by comparing frequencies of loss for different locus in low-grade\r\ndysplasia with known outcome, i.e. low-grade lesions that did not progress into\r\ncancer with morphologically similar lesions that did develop into SCC.\r\nIn summary, the data suggest that at least 3 regions of alteration are present on 11q in\r\nboth oral premalignant and malignant lesions and that 1 of these regions, identified as\r\ncontaining the novel marker D11S4207, might play a significant role in the early\r\ndevelopment of the disease. The data further support the use of these markers to identify\r\nprogression risk for early oral premalignant lesions.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/12316?expand=metadata","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/extent":[{"value":"7321674 bytes","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/elements\/1.1\/format":[{"value":"application\/pdf","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"A L L E L I C I M B A L A N C E AT 11Q IN O R A L CANCER AND PREM ALIGN ANT LESIONS by DING A N B.Sc, Lanzhou University, China, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R Of SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Oral Biological and Medical Sciences; The Faculty of Dentistry) We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH C O L U M B I A May 2002 \u00a9 D i n g A n , 2002 ALLEL IC I M B A L A N C E AT 11Q IN O R A L CANCER AND PREMALIGNANT LESIONS by DING AN B.Sc, Lanzhou University, China, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R Of SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Oral Biological and Medical Sciences; The Faculty of Dentistry) We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH C O L U M B I A May 2002 \u00a9 D i n g A n , 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2\/88) A B S T R A C T Oral squamous carcinomas (SCC) remain a significant public health problem worldwide despite advances in therapy and local disease control. New innovative strategies must be developed for the prevention, early detection and treatment of oral carcinoma. Such approaches w i l l be heavily dependent on a better understanding of the molecular mechanisms underlying carcinogenesis at this site. This thesis describes a series of studies done on oral cancers and premalignant lesions to better define the role of alterations on chromosome 11 in the development of oral cancer at this site. Although numerous studies have reported the presence of alterations on this chromosome arm in oral cancers, few studies have examined premalignant lesions in order to determine whether such alterations play a role in the development of the disease. The objectives of this thesis were: 1) to use microsatellite analysis to examine D N A extracted from severe dysplasia, carcinoma in situ (CIS) and S C C for novel alterations on 1 l q ; 2) to determine at what stage of oral cancer development the alteration occurred by performing microsatellite analysis on a spectrum of stages of oral premalignant lesions (hyperplasia, mi ld dysplasia, moderate dysplasia, severe dysplasia, CIS) as wel l as invasive S C C . The data obtained was compared to 2 previously studied hotspots on 1 l q : the int2 (1 l q l 3 ) and D11S1778 (1 lq22-23); and 3) to determine the significance of allelic imbalance (Al ) at int2 (1 l q l 3 ) and D11S1778 (1 lq22-23) to the progression of i i oral premalignant lesions by comparing frequencies of loss for different locus in low-grade dysplasia with_known outcome, i.e. low-grade lesions that did not progress into cancer with morphologically similar lesions that did develop into S C C . In summary, the data suggest that at least 3 regions o f alteration are present on 1 l q in both oral premalignant and malignant lesions and that 1 of these regions, identified as containing the novel marker D11S4207, might play a significant role in the early development of the disease. The data further support the use o f these markers to identify progression risk for early oral premalignant lesions. i i i T A B L E OF CONTENTS A B S T R A C T ii T A B L E O F C O N T E N T S iv LIST O F T A B L E S ix LIST O F FIGURES xi LIST O F ABBREVIATIONS xii A C K N O W L E D G E M E N T S xiv DEDICATION xv I. INTRODUCTION 1 1.1. Overview 1 1.2. Cl inical , histological and molecular alterations during oral carcinogenesis 4 1.2.1. Cl inical alterations during oral carcinogenesis 4 1.2.2. Histological alterations during oral carcinogenesis 7 1.2.2.a. Normal oral mucosa 7 1.2.2.b. Oral premalignant lesions and squamous cell carcinoma (SCC) 8 iv 1.2.3. Molecular alterations during oral carcinogenesis 12 1.2.3.a. The importance of studying genetic changes in cancer progression 12 I.2.3.b. Oncogenes and tumor suppressor genes (TSGs) 14 1.2.3.c. Oncogenes and TSGs in oral premalignant and malignant lesions 18 1.2.3.d. Microsatellite analysis amd loss of heterozygosity (LOH) studies 19 1.2.3 .e. Microsatellite analysis of oral cancer and premalignant lesions 24 1.2.3.f. Genetic progression model for head and neck cancer 25 1.2.3 .g. Prediction of risk of progression for oral premalignant lesions (OPL) 29 1.3.1. Genetic changes on 11 q in a variety of cancers 31 1.3.2. Genetic alterations at 1 l q in head and neck squamous cell carcinoma ( H N S C C ) 34 1.3.3. Mechanism of alteration at 1 l q l 3 and 1 lq22-23 36 1.3.4. 11 q alterations in head and neck premalignant lesions 43 1.3.5. M a i n genes on chromosome 1 l q 45 I.3.5.a. CCND1 {cyclin Dl, PRAD1) gene at 1 lq l3 48 I.3.5.b. int2 (FGF3) gene at 1 l q l 3 49 I.3.5.C. HST1 gene at 11 ql3 50 I.3.5.d. EMS1 gene at l l q l 3 51 v 1.3.5.e. MEN I gene at 1 l q l 3 52 I.3.5.f. RIN1 gene at l l q l 3 52 I.3.5.g. y47Mgeneat llq23.1 53 II. STATEMENT OF PROBLEMS 56 II. 1. Where are the additional tumor genes at 1 l q l 3 located? 56 II.2. A t what stages of oral cancer development does A I occur for int2 (1 l q l 3 ) , D11S1778 (1 lq22.3) and any novel loci identified in this study? 57 III. OBJECTIVES 59 IV. HYPOTHESES 60 V. MATERIALS AND METHODS 61 V . 1. S ample collection 61 V . 2 . Sample sets 61 V . 3 . Diagnostic criteria for the samples , 64 V.4 . Cl inical information 64 V . 5 . Slide preparation 65 V .6 . Microdissection ...66 V . 7 . Sample digestion and D N A extraction 66 v i V . 8 . D N A quantification 67 V . 9 . Primer extension preamplification (PEP) 67 V.10. Coding samples 68 V . l l . End-Labeling 68 V.12. Microsatellite analysis: P C R amplification 70 V . 13. Scoring of allelic imbalance 71 VI. R E S U L T S 72 V I . 1. Choice of microsatellite markers for this study 72 VI.2. A l at D11S4207 in oral SCCs and premalignant lesions 75 VI .3 . The timing of induction of A l at D11S4207, int2 and D11S1778 during histological progression 81 VI.4. Further evidence in support of the A l at D11S4207 being an independent event. 85 VI .5 . Fine-mapping at D11S4207 87 VI.6. Al le l i c imbalance at chromosome 1 l q and malignant progression risk 88 VII. DISCUSSION 93 V I I . l . Al le l i c imbalance of genes at D11S4207, D11S1778 and int2-cycline Dl during v i i multistage oral carcinogenesis 93 VII . 1.1. Temporal changes of the 3 loci at 11 q 94 VII . 1.2. Significance of the changes of the 3 loci at 1 l q 96 VII. 1.2.1. Significance of AI at Dl 1S4207 96 VII.1.2.2. Significance of AI at DUS1778 (1 lq22-23) and int2 (M2-cyclin Dl region) 99 VII.2. Al le l i c imbalance at D11S1778 and int2 is associated with cancer risk 100 VII.3. D11S4207, a new hot spot at 1 l q l 3 102 VII.4. Ending Mark 104 BIBIOGRAPHY 105 v i i i LIST OF T A B L E S Table 1. A l in oral lesions 26 Table 2. A l on 1 l q in various cancers as detected by microsatellite analysis 33 Table 3. A l at loci on 1 l q in H N S C C s . 35 Table 4. Amplification at 1 l q l 3 as identified with F I S H in H N S C C 41 Table 5. Genetic alterations at 1 l q of head and neck premalignant lesions 44 Table 6. List o f genes located at 1 l q l 3 46 Table 7. Histological groups in sample set 1 62 Table 8. Histological groups in sample set 2: the progression test series 63 Table 9. Al le l i c imbalance at D11S4207 in a spectrum of primary lesions with different histological diagnoses 77 Table 10. Al le l i c imbalance at D11S4207, int2 and D11S1778 oral premalignant lesions and S C C 84 Table 11. Patterns of alteration at Dl 1S4207 and int2: frequencies at which these alterations occur together or independent of each other 86 Table 12. Patterns of alteration at D11S4207 and D11S1778: frequencies at which these alterations occur together or independent of each other 87 ix Table 13. Demographic information of patients with low-grade dysplasia 90 Table 14. Al le l i c imbalance of DllSI778 (ATM) and int2 (intl-cyclin Dl) in progressing and non-progressing hyperplasia and low-grade dysplasia 91 Table 15. Association of allelic imbalance at D11S4207 and 3p &\/or 9p in low-grade dysplasias 99 LIST OF FIGURES Figure 1. Clinical presentation of oral premalignant lesions 5 Figure 2. Histological progression model for oral premalignant and malignant lesions. 11 Figure 3. Schematic illustration o f microsatellite analysis and A I 23 Figure 4. Molecular progression model of oral cancer proposed by Rosin and Zhang (2001) 28 Figure 5. The procedures of B F B cycle model of 1 l q l 3 amplification 38 Figure 6. Microsatellite results for cases with A I at D11S4207 and retention at int2 74 Figure 7. Microsatellite analysis of S C C cases at D11S4207 and int2 76 Figure 8. Microsatellite analysis of hyperplasia cases at D11S4207 79 Figure 9. Comparison of A I frequencies observed at D11S4207 with those at 3pl4 , 9p21 and 8p 80 Figure 10. Comparison of A I frequencies observed at D11S4207 with those seen with microsatellite markers at D11S1778 and mt2 83 Figure 11. Probability o f having no progression to cancer, according to A I at Dl 1S1778 or at int2 92 x i LIST OF ABBREVIATIONS A I allelic imbalance ATM ataxia telangiectasia mutated B A C bacterial artificial chromosome B F B Breakage-fusion-bridge CCND1 cyclin D l CIS carcinoma in situ dmin double-minute chromosomes D N A deoxyribonucleic acid FGF fibroblast growth factor F I S H fluorescence in situ hybridization H & E hematoxylin and eosin hsrs homogeneously staining regions H N S C C head and neck squamous cell carcinoma HST1 heparin secretory transforming factor 1 L O H loss of heterozygosity M b megabase pair MEN1 multiple endocrine neoplasia type 1 O P L oral premalignant lesion P C R polymerase chain reaction S C C squamous cell carcinoma T S G tumor suppressor gene W H O World Health Organization x i i i ACKNOWLEDGEMENTS I would like to take this opportunity to thank my supervisor Dr. Lewei Zhang for her support and guidance throughout my degree; and to extend my warmest gratitude to my co-supervisor Dr. Mi r i am P. Rosin for her academic help and constant moral support. I am also grateful to Dr. Douglas Waterfield for being my committee member and for his valuable comments. I really appreciate technicians and other students in Dr. Rosin and Dr. Zhang's lab for their support. I also would like to take this opportunity to thank my grandma, my uncle Y u and his wife Huijun, my brother Fan and my cousin Ruobing for their continuous support and understanding. x i v DEDICATION To My Mom and My Dad Who Love Me and Whom I Love Forever I. INTRODUCTION 1.1. Overview Oral cancer is the sixth most common cancer in the world, accounting for about 3% of all new cancers in Western countries (Harras et al, 1996; Greenlee et al, 2001) and up to 40% of all cancers in places such as India (Saranath et al, 1993). Despite refinement of surgical techniques and adjuvant therapies, the prognosis of oral cancer has remained unchanged for decades with 5-year survival rates of 40-50% in the Western world and even lower in undeveloped countries (20-43%) (Rao and Krishnamurthy, 1998; Greenlee et al, 2001). The major cause of this high mortality rate is the late-stage at which most cancers are identified, resulting in a high local recurrence and formation of second primary malignancies even in successfully treated cases (Cooper et al, 1989; Day and Blot, 1992; Lippman and Hong, 1989; Shikhani et al, 1986). The development of novel approaches for the prevention, early detection, and effective treatment of this disease are critical to improving outcome. Such approaches are dependent on a better understanding of the molecular and cellular mechanisms underlying the tumorigenesis process. Molecular analyses of oral cavity tumors have uncovered a number of recurrent genetic events that appear to underlie the development of the disease. Among these changes are frequent alterations to loci (and genes) on chromosome 1 l q . Microsatellite analysis of 1 oral tumors suggests the presence of at least 2 regions on this chromosomal arm containing putative TSGs or oncogenes. These regions include 1 l q l 3 and 1 lq22-23, each of which contain loci that show frequent allelic imbalance in tumors, with such alteration occurring in 30-80% of cases (Dwight et al, 2000; Ah-See et al, 1994; Venugopalan et al, 1998; D ' A d d a et al, 1999; Uzawa et al, 1996; Lazar et al, 1998; H u i et al, 1996). The region around 1 l q l 3 is of particular interest in that amplification of this region (as identified with fluorescent in situ hybridization or FISH) has been reported in 30-50% of oral cancers (Fujii et al, 2001; Ott et al, 2002; Bayerllein et al, 2000; A lav i et al, 1999; Will iams et al, 1993; Mul ler et al, 1997; Meredith et al, 1995; Fortin et al, 1997). Several genes potentially important for creating the malignant phenotype are located at l l q l 3 or l lq22-23 including int2, FGF4, EMS1, RIN1, MEN1, PAK1, ATM and the most commonly reported gene, cyclin Dl. However, the evidence in support of the involvement of the majority of these genes is still limited. There is little information on the involvement of alterations on 1 l q in oral premalignant lesions. Such alterations have been reported as being present in head and neck premalignant lesions (El-Naggar et al, 1995; Califano et al., 1996; Poh et al, 2001). However, the numbers of cases used in these studies is small and the association with the degree o f dysplasia is reported in only one of these studies (Poh et al, 2001). That study suggested that 1 l q alterations may be occurring late in carcinogenesis, between severe dysplasia and S C C (Poh et al, 2001). There is also some preliminary evidence from this laboratory that suggests that allelic imbalance at 1 l q is associated with an increased risk for progression (Rosin et al, 2000). These studies need to be confirmed using a larger 2 number of cases and expanding the analysis to other loci on this arm in order to better identify both the sites containing genes of importance to progression o f oral premalignant lesions and to determine whether or not such markers can be used as indicators of risk of progression for early lesions. The goal of this thesis was to examine oral cancers and premalignant lesions for alteration to markers on the 1 l q arm in order to better define the timing and frequency of alterations at 1 l q . A second goal was to determine whether there was an association of alteration at 1 l q with clinical information and prognosis. A final objective was to use microsatellite analysis to begin to define novel regions of alteration in oral premalignant lesions that could later be studied for candidate tumor genes located at 1 l q . The following introduction begins with a summary of the clinical and histological alterations that occur during oral carcinogenesis and the limitations of histopathological criteria in identifying outcome for lesions with no dysplasia or with low-grade dysplasia. A brief summary of genetic alterations associated with oral tumorigenesis is then given followed by a more specific presentation of 1 l q alterations and cancer, with an emphasis on oral squamous cell carcinomas and premalignant lesions. 3 1.2. Clinical, histological and molecular alterations during oral carcinogenesis 1.2.1. Clinical alterations during oral carcinogenesis A premalignant lesion, as defined by the World Health Organization (WHO) , is 'morphologically altered tissue in which cancer is more l ikely to occur than in its apparently normal counterpart' ( W H O , 1978). This definition was reaffirmed in 1994 by the International Collaborative Group on Oral White Lesions (Axe l l et al, 1996). In the oral cavity, premalignant lesions most frequently present as leukoplakia and erythroplakia (Axe l l et al, 1996). Leukoplakia (Figure 1, A ) means \"white patch\". It occurs on membranes such as the mucosa o f the oropharynx, larynx, esophagus, and genital tract. The term leukoplakia should not be applied to all the white patches of these mucosal surfaces. The World Health Organization (1978) defines leukoplakia in the oral cavity as a white patch or plaque of oral mucosa that cannot be characterized clinically or pathologically as any other diagnosable disease and is not removed by rubbing. Usually, a definitive diagnosis of oral leukoplakia is made as a result of the identification, and i f possible, elimination of suspected etiological factors (Axel l et al, 1996). Likewise, the W H O defines erythroplakia (Figure 1, B) as a fiery red patch that cannot 4 be characterized clinically as any other definable lesion (Pindborg et al, W H O , 1997). In contrast to leukoplakia which are often benign in nature, lesions with erythroplakia are generally either already malignant or are at high-risk for transformation to malignancy (Bouguot and Ephros, 1995; Mashberg, 1977; Waldron and Shafer, 1975). It is generally believed that a premalignant lesion that most frequently presents clinically as leukoplakia or erythroplakia often precedes oral cancer. However, the majority o f leukoplakias w i l l not progress to cancer. Figure 1. Clinical presentation of oral premalignant lesions A B A : Leukoplakia on right latero-ventral tongue; B : Erythroplakia on right latero-ventral tongue. A number of clinical factors have been found to affect the malignant risk of oral premalignant lesions. These include the appearance, size, site and duration o f the lesion, the consumption of tobacco and alcohol, and history of head and neck cancer, or of Candida infection (van der Waal et al, 1997). The following is a brief discussion of these attributes. Leukoplakia can be classified as either homogeneous or non-homogeneous, according to 5 the clinical appearance of oral premalignant lesions. Homogeneous leukoplakias are those lesions showing a consistent color and texture; these lesions have a lower risk for malignant transformation compared to non-homogeneous leukoplakia (Axe l l et al, 1996; Pindborg et al, W H O , 1997). Leukoplakia occurs throughout the oral cavity, with those in the buccal and mandibular sites being the most common. Leukoplakia located on the floor of the mouth, ventrolateral surface of the tongue and soft palate have an increased cancer risk (Schell and Schonberger, 1987; Mashberg and Meyers, 1976). Hence, these regions are called high-risk areas whereas the other oral sites are called low-risk areas (Schell and Schonberger, 1987; Mashberg and Meyers, 1976). Cl inical risk factors help clinicians decide whether an oral lesion should be biopsied for histopathological evaluation and also help in the overall judgment of malignant risk of the lesions. However, these clinical factors have a limited ability to precisely predict cancer risk. Currently, histological criteria represent the gold standard forjudging the risk o f malignant transformation for oral premalignant lesions. 6 1.2.2. Histological alterations during oral carcinogenesis 1.2.2.a. Normal oral mucosa The oral cavity is lined by oral mucosa, which consists of stratified squamous epithelium and a layer of connective tissue called the lamina propria. The stratified squamous epithelium can be divided largely into basal and prickle cells. The lamina propria lies underneath the epithelium and contains blood vessels and lymphatic vessels, small nerves, fibroblasts, collagen and elastic fibers. It functions to nourish and support the epithelium. At the boundary of the epithelium and connective tissue is a layer of basal cells. Some of these basal cells are \"stem cells\" of the tissue which possess the ability to replicate themselves to replace cells lost, during tissue turnover. When a basal \"stem\" cell divides, it may give rise to new basal cells or differentiate to form the larger polyhedral-shaped prickle cells. A s the prickle cells mature, they push towards the surface where they shrink in size, become long and flat, and are eventually desquamated. Oral epithelium is usually non-keratinized except for mucosa lining the attached gingiva, hard palate, dorsal surface of the tongue, and lips. The majority o f oral premalignant and malignant lesions arise from this stratified squamous epithelium of oral mucosa and the malignant tumors are called squamous cell carcinoma (SCC). 7 I.2.2.b. Oral premalisnant lesions and squamous cell carcinoma (SCC) Oral S C C is believed to evolve from normal tissue that develops premalignant lesions with these lesions increasing in severity and eventually becoming malignant. To assess the risk of malignant transformation of leukoplakia or erythroplakia, a biopsy is taken of the clinical lesion and examined for the presence and degree of dysplasia. The World Health Organization has established the following criteria for histological diagnosis of oral dysplasia (1978): 1. Loss of polarity of the basal cells 2. The presence o f more than one layer having a basaloid appearance 3. Increased nuclear\/cytoplasm ratio 4. Drop-shaped rete-ridges 5. Irregular epithelial stratification 6. Increased numbers and abnormality of mitotic figures 7. The presence o f mitotic figures in the superficial half o f the epithelium 8. Cellular pleomorphism (variation in shape and size) 9. Nuclear hyperchromatism (dark staining nuclei) 10. Enlarged nucleoli 11. Loss of intercellular adherence 12. Keratinization of single cells or cell groups in the prickle cell layer 8 Architecturally, dysplastic lesions are further divided into mild, moderate, and severe forms depending upon how much of the tissue is dysplastic. M i l d dysplasia is a lesion in which the dysplastic cells are confined to the lower one third of the epithelium. Moderate dysplasia is a lesion in which the dysplastic cells are evident in about half the thickness of the epithelium. Severe dysplasia is a lesion in which the dysplastic cells have filled the lower two-thirds of the epithelial thickness. In carcinoma in situ (CIS), the dysplastic cells occupy the entire thickness of the epithelium (bottom to top changes) although the basement membrane is still intact (Lumerman et al, 1995). Invasion of dysplastic cells through the basement membrane into the underlying stroma and\/or the dissemination of these cells to other sites through lymphoid and circulatory systems are events associated with development of invasive S C C . A histological progression model has been established for oral cancer (Figure 2). In this model, oral cancers progress through hyperplasia and increasing degree o f dysplasia (mild, moderate and severe) to CIS, and finally break through the basement membrane and become SCCs . Severe dysplasia and CIS usually are grouped together as high-grade dysplasia, because both are late stage, preinvasive lesions and the distinction between them is often difficult and does not appear to be of practical value in the management of oral mucosa (Pindborg et al, W H O , 1997). The presence and the degree of dysplasia are believed to have a huge impact on the malignant risk of the premalignant lesions. A l l studies to date have shown that leukoplakia with dysplasia have higher malignant risk than those without dysplasia 9 (Waldron and Shafer, 1975; Lumerman et al, 1995). A large clinical study by Silverman et al. (1984) found that during a mean follow-up period of 7.2 years, more than 36% of leukoplakia lesions with epithelial dysplastic features eventually underwent malignant transformation, whereas those leukoplakia without dysplasia only demonstrated a malignant rate o f 15%. The relationship of the malignant risk and degree o f dysplasia is further demonstrated by parallel studies from uterine cervix and other systems and organs including skin and respiratory system (Boone et al, 1992; Braithwaite and Rabbitts, 1999; Geboes, 2000; Pinto and Crum, 2000; Shekhar et al, 1998). A s a result, the gold standard forjudging the malignant potential of premalignant lesions in these organs and systems, including the oral cavity, is the presence and degree of dysplasia. It is important to note, however, that there are limitations in the use of histological criteria to predict malignant risk of oral premalignant lesions. These criteria have a good predictive value for high-grade dysplasia, severe dysplasia and CIS, which have a high chance of progressing into invasive lesions (Banoczy and Csiba, 1976; Schepman et al, 1998). However, histology is a poorer predictor o f malignant risk for low-grade dysplasia (mild and moderate dysplasia). Low-grade lesions that ultimately progress to tumors appear histologically mimic to those that regress or remain unchanged over extended period of time. Since such lesions represent the majority o f premalignant lesions, a way to differentiate progressing from non-progressing would have a large impact on prevention of oral cancer. Advances in molecular techniques may provide a new direction to solving this problem. 10 X H 1.2.3. Molecular alterations during oral carcinogenesis 1.2.3.a. The importance of studying genetic changes in cancer progression It is now widely accept that cancer develops through a series of genetic events that parallel the histopathological progression of a lesion through premalignant stages to carcinoma in situ, and, finally, into an invasive lesion. Molecular progression models outlining these genetic changes were established in colon cancer first (Vogelstein et al., 1988), and then developed for many other solid tumors. According to these models, different genetic changes occur in different stages of the disease process and play an integral role in the progression of the lesion to cancer. Many of these genetic events take place well before a given tumor produces clinical symptoms and often before a benign lesion or focus of dysplasia develops into an invasive cancer (Sidransky, 1997). This suggests that i f such changes could be used to distinguish lesions with an elevated risk of developing into cancer. A s an example of such an approach, there is substantial evidence that microsatellite analysis can be used to predict outcome for oral premalignant lesions (Califano et al, 1996; Partridge et al., 1998; 1999). In a recent report from this laboratory, it was shown that loss of heterozygosity (LOH) at 3p and\/or 9p was one o f the earliest changes associated with an increased risk of malignant transformation (Rosin et al., 2000). Low-grade dysplasia with L O H at 3p and\/or 9p had approximately 4 times the risk of 12 developing into cancer compared to those without such loss. Cases that had L O H at 3p and\/or 9p with additional loss on other chromosome arms (4q, 8p, 1 l q , or 17p) had a 34-fold increase in relative cancer risk. These results are supported by a number of studies from other laboratories (Partridge et al, 1998; 1999; Lippman and Hong, 2001) and strongly support the use of molecular markers to predict cancer risk o f premalignant lesions. This article w i l l be discussed in more detail in section 1.2.3.f. Genetic alterations have also been found to correlate with clinical parameters of cancer, such as recurrence, metastasis and poor prognosis. This has been observed for many types of cancer. For example, L O H of loci at 3p25.1, 13ql2 or 17pl3.3 have been associated with lymph node metastasis and poor prognosis of breast cancers (Hirano et al, 2001). Deletion at 3p25.3-p23 has been shown to be associated with metastasis of endocrine pancreatic carcinoma (Barghorn et al., 2001). A putative tumor suppressor gene (TSG) located at 20pl 1.23-pl2 has been reported to be involved in the development of prostate cancer metastases (Goodarzi et al., 2001). L O H at 5q21-22 has been reported to be linked to known oral S C C etiologic factor (human papilloma virus) and the prognosis of patients (Mao et al, 1998). Another study has concluded that microsatellite analysis for 18q21.1 and 9p21-22 is capable of predicting the clinical outcome of bladder cancer patients (Uchida et al, 2000). A l l these papers have shown that genetic alterations can be used to predict clinical outcomes and help to decide treatment. Another reason for studying genetic alterations is to facilitate the development of new regimen such as treatment of gene therapy for premalignant lesion and cancers. The 13 efforts in cancer gene therapy focus on four types of approaches: chemogene therapy, such as the introduction of genes that confer susceptibility to chemotherapeutics; immunogene therapy, which involves modulation of the patient's immune response capacity; tumor suppressor gene-mediated inhibition of tumor growth promoters like oncogenes and cytokines; and inhibitors of tumor angiogenesis and invasiveness (Vogelstein and Kinzler , 1998). Each of these approaches requires a basic understanding o f the genetic background o f a patient's lesion. For example, in gene therapy, the transfer of TSGs to cancer cells to suppress tumorigenesis requires the identification of such genes in a patient's lesion. A final example o f how studies of genes altered in tumorigenesis can be used to tailor intervention and treatment strategies is exemplified by carriers o f APC (adenomatous polyposis coli) mutations. Such individuals have a higher risk of developing colon cancer (Nagase et al, 1992; Miyoshi et al, 1992; Powell et al, 1992; Smith et al, 1993). In such situations, a prophylactic colorectomy is used to reduce the incidence of this disease dramatically (Tomlinson et al, 1997). I.2.3.b. Oncogenes and tumor suppressor genes (TSGs) Nowel l proposed in 1976 that neoplastic transformation occurred in a single cell that had a critical genetic alteration that gives it a growth advantage over its neighboring cells (Nowell, 1976). This was followed by successive rounds of mutation and expression 14 with the accumulation of multiple genetic alterations during the progression of the disease. In the head and neck region, 7-10 independent genetic events are believed to be involved in the production of invasive S C C (Renan, 1993). This theory has been supported by a genetic progression model for head and neck cancer developing by Califano et al. (1996). Based on this model, specific genetic alterations occur during the progression from normal mucosa through increasing degrees o f dysplasia and finally to invasive S C C . These genetic events include the mutation of a number of critical genes that control the processes of cellular proliferation and differentiation in a tissue. The genes associated with carcinogenesis are classified into two main groups: oncogenes and tumor suppressor genes (TSGs). Oncogenes, which originally were identified as the transforming genes in viruses, are altered forms o f normal cellular genes called proto-oncogenes. In human cancers, proto-oncogenes are frequently located adjacent to chromosomal breakpoints and are targets for mutation. The products of proto-oncogenes play a key role in regulating the cascade of events that maintains the ordered processes through the cell cycle, cell division, and differentiation. More than 50 different proto-oncogenes have been identified, coding for proteins that function as growth factors, growth factor receptors, cytoplasmic second messengers, protein kinases, nuclear phosphoproteins, transcription factors and others. They can be roughly subdivided into two groups. One class of genes rescues cells from senescence and programmed cell death, acting as immortalizing genes. A second class of genes reduces growth factor requirements and induces changes in cell shape that results in a continuous proliferative response (Vogelstein and Kinzler , 1998). 15 A number of mechanisms have been described for the mutation o f these proto-oncogenes to oncogene (called activation) including point mutations, gene amplification and chromosome translocations (Vogelstein and Kinzler, 1998). These alterations can involve the coding region of the gene, resulting in alteration of structure and activity of coded proteins. Alternatively they can alter the regulating region o f a gene, resulting in the inappropriate expression of the gene. This resulting genetic alteration is autosomal dominant, meaning that only one o f the two gene copies needs to be changed for an effect to be observed. The mechanism of gene amplification requires further comment at this time, since the region to be studied in this thesis (1 l q l 3 ) has often been reported as being amplified during carcinogenesis. The term gene amplification refers to an increase in copy number of a gene or a specific, subchromosomal region. Gene amplification almost always results in the overexpression of one or more genes contained on the amplicon. The amplification and consequent overexpression of a number of genes can act to confer a selective advantage on a cancer cell. In contrast to oncogenes, tumor suppressor genes (TSGs) are a group o f genes encoding proteins, which, through a variety of mechanisms, function to negatively regulate cell growth and differentiation pathways. The functions of TSGs must be lost in order for tumorigenesis to occur. Loss of a T S G ' s function requires the inactivation of both gene copies (maternal and paternal) through mutation, deletion, or other mechanisms such as methylation (Knudson et al., 1977, 1985, 1993). The most commonly reported process of 16 T S G loss in sporadic tumors involves the inactivation of a gene by a point mutation followed by inactivation of the second copy by any one of many mechanisms including deletion. A commonly used procedure to identify such deletion is the microsatellite assay, to be discussed in a later section. This assay is the primary techniques used in this thesis. Many tumor suppressor genes have been localized and identified, including p53, KB (retinoblastoma), VHL (the gene responsible for von Hippel-Lindau syndrome), FHIT (fragile histidine trial), pi6, DPC4, APC (adenomatous polyposis coli), doc-1 (deleted in oral cancer), TSC2, BRCA1, NF-1, NF-2 and WT-1 (Mao et al, 1996; Reed et al, 1996; Gleich et al, 1996; Todd et al, 1995; Largey et al, 1994; Pavelic et al, 1997; Uzawa et al, 1994; Mao et al, 1996; and K i m et al, 1996; Lat i f et al, 1993; Kanno et al, 1994; Sparks et al, 1998). Although the cellular functions of tumor suppressor proteins, such as p l 0 5 - R B , p53 and p l 6 , are becoming increasingly well understood, others remain largely undefined. It is clear, however, that the tumor suppressor proteins w i l l exhibit a variety of functions within the cell. Functional loss o f TSGs is one o f the most common genetic alterations during carcinogenesis including that of the head and neck region. Thus, the defining of chromosomal regions that harbor biologically important suppressor genes may have broad practical implications not only on tumor progression, but also on the clinical management of cancers and premalignant lesions. 17 I.2.3.C. Oncogenes and TSGs in oral premalisnant and malignant lesions Few oncogenes have been identified as showing mutation in head and neck or oral squamous cell carcinoma, although changes in the expression of many potential oncogenes has been reported. These genes include ras, cyclin-Dl, myc, erbB, bcl-l, int2, CK8 and CK19 (Kiaris et al, 1995; Lese et al, 1995; Saranath et al, 1993; Warnakulasuriya et al, 1992; Wong et al, 1993; Bartkova et al, 1995; X u et al, 1995; Masuda et al, 1996; Riviere et al, 1990). Our knowledge about the frequency of mutation of these genes in different populations is still somewhat limited. For example, ras and myc mutations appear to be more prevalent in head and neck tumors occurring in the Far East, possibly due to the use of chewing tobacco and betel quid by these populations (Anderson et al, 1994; Clark et al, 1993; Paterson et al, 1996; Saranath et al, 1993). Mutations of K-ras have been identified in approximately 35% of tumors in the latter group; however, the prevalence of these mutations in Western patients is only five per cent (Kiaris et al, 1995; Matsuda et al, 1996; Sakata, 1996). In addition, very few studies have included an analysis of mutation frequencies in premalignant lesions. The few studies available tend to use immunohistochemical analysis and look at increased expression o f the gene, not mutation. For example, H o u et al (1992) reported a progressive increase in c-erb-2\/neu expression as premalignant lesions advanced to malignant lesions. However, it is not known whether this effect was due to a mutation of the gene itself or to a dysregulation of the expression of this gene resulting from a downstream effect of another mutation. 18 On the other hand, many studies have focused on the role of TSGs in oral carcinogenesis. Some of the TSGs involved in head and neck cancers include p53, Rb (retinoblastoma), andpl6INK4A (Gallo et al, 1999; Gleich et al, 1996; Jares et al, 1999; Liggett et al, 1996; Papadimitrakopoulou et al, 1997; Partridge et al, 1998 and 1999a; Pavelic and Gluckman, 1997; Reed et al, 1996; Sartor et al, 1999). Other potential candidates are FHIT (Fragile histidine triad), APC (adenomatous polyposis coli), doc-1 (deleted in oral cancer), VHL (the gene responsible for von Hippel-Lindau syndrome) and TfiR-II (the gene coding for transforming growth factor type II receptor) (Croce et al, 1999; Largey et al, 1994; Mao et al, 1996c; Mao, 1998; Todd et al, 1995; Uzawa et al, 1994; Waber etal, 1996). Recent advancement in molecular analysis techniques has rapidly revolutionized the ability to look at these genetic alterations. Most studies on TSGs , particularly those in oral premalignant lesions, use microsatellite analysis to identify loss of heterozygosity (LOH) in D N A extracted from epithelial cells belonging to these lesions. This is also the major technique used to conduct research for this thesis. I.2.3.d. Microsatellite analysis amd loss of heterozygosity (LOH) studies Microsatellite analysis is a powerful molecular technique for identifying and studying TSGs. It can detect changes from as small as a few thousand nucleotides to as large as a whole chromosome in one of a pair o f chromosomes. The assay is designed to assess 19 alterations to polymorphic chromosomal regions that map close to or within putative or known TSGs . Two methods are available for the study of L O H : restriction fragment length polymorphism (RFLP) analysis and microsatellite analysis. The advantages of using microsatellite analysis are twofold. First, microsatellite repeat markers are highly polymorphic and wel l distributed throughout the human genome. They show levels o f heterozygosity between 30-80%, significantly above the level observed with the R F L P analysis based on base substitutions at endonuclease recognition sites. Second, this [y-p 3 2 ] end-labeled PCR-based approach is much more sensitive than R F L P analysis and requiring only a small amount of D N A (5 nanograms or less per reaction). This aspect is critical for the study of premalignant lesions given the small size of such lesions. Another advantage is that microsatellite makers can be used on D N A extracted from paraffin-embedded archival samples in addition to fresh or frozen samples. This allows one to use the large number of specimens in hospital archives to perform retrospective analysis of lesional progression. The following is a brief description of this procedure. Microsatellites contain runs of short and tandemly repeated sequences o f di- , tri-, or tetra-nucleotides, such as - G T G T G T - , - G T A G T A G T A - , o r - G T A C G T A C G T A - (Figure 3). The number of such tandem repeats is highly polymorphic in the population. Thus, individuals often contain a different number of copies of the repeat in maternal and paternal alleles. These regions are well interspersed throughout the human genome (e.g., estimated every 30-60 kilobase pairs (kb) for C A repeats) and are highly conserved 20 through successive generations (Ah-See et al, 1994; Beckman and Weber, 1992). The assay involves amplification of region containing microsatellites with polymerase chain reactions (PCR) using radioactively labeled nucleotides. The reaction products are separated on polyacrylamide gel, with separate bands being produced whenever on individual is heterozygous for a region (i.e., when paternal and maternal copies of the region contain different numbers of repeats). The intensity o f the two alleles is compared for D N A isolated from normal and tumor or dysplasia cells in a tissue. A n alteration in intensity of the bands in the lesion D N A compared to normal D N A is scored as L O H (see section V.13. for details). The observation o f a high frequency o f L O H in a microsatellite region in a set o f tumors or dysplasias is suggestive of the presence of a putative suppressor gene nearby. However, it is important to note that microsatellite analysis cannot distinguish between duplication or low-level amplification of an allele and deletion of an allele, particularly i f there are contaminating normal cells within the tumor (Ah-see et al, 1994). Both alterations would result in a change in the relative signal intensities for the two alleles in the lesion D N A . To determine whether a deletion or amplification has occurred, other assays are used. One approach involves the use of fluorescence in situ hybridization (FISH) on cell or tissue preparations to determine the number copies of the region in cells. Alternatively, Southern blots could be used with specific gene probes. Both procedures are time-consuming. The latter procedure requires significantly more D N A than the microsatellite analysis. Other procedures involve sequencing the gene to 21 determine whether a mutation exists. This process also requires large quantities of D N A o r R N A . Because of the difficulty in differentiating amplification from deletion as a source of alteration to microsatellite signals, many investigators use the term allelic imbalance (Al ) to describe the change o f signal intensity observed with microsatellite analysis. I w i l l use both terms indiscriminately throughout this thesis. 22 T3 C \u00ab \u2022 \u2014 \u00ab S ea \"3 o s-u 1 C M O a _o +3 S3 i-\u2022M C O 3 U a E 2 I - I c3 +-\u00bb .5 CD -si < 8 o O J 3 a CD o \"\u20223 0) 'SH o o \u2022 H O CD t-i S O v U a o cii a \u00ab ;=s o 5 a 3 & 2 -s ^ 2 ^ CD a PQ \u00b0 2 C O C 4 H O O a o P o J3 o o -4-* (3 (D O I ! 3 CD CD O CD S3 O o l-l u w T3 co J 3 3 H S3 . . CD m $ CD X > b0 S3 \"\u00a7 o \u00b0 Q g p J 3 a 2 3 O PQ U PQ o C O CD ^ o cd CD 3 o o .a I C M o C O CD ( 3 o CJ CJ o o CD 7L M O \u00bb - i d CD h-1 T 3 CD \u2022a cd a cs 3 , c cj u IS a E O S u PQ to PQ 3 .2 CO 3 11 a o T3 S3 C3 ^! pa 0 0 en CD ha S T3 cu U o u OH CU JS H CU s DC cu m o m According to Shuster's study, these two fragile sites play critical roles in initiating the B F B cycle to cause the amplification of 1 l q l 3 , as well as in determining the size and genetic content of amplified units. One possible fragile site involved in gene amplification of 1 l q l 3 in oral cancer, FRA11A, is found to be located between RIN1 and CCND1 (Shuster et al, 2000). Further evidence to support the occurrence of a second break distal to CCND1 is the observation that 1 l q l 3 amplification is usually accompanied by distal deletions between Y A C s 55G7 and 749G2 (Jin et al., 1998). Since the genes in an amplicon always amplify together, some believe that samples with int2 amplification should reflect the amplification of other closely localized genes. Consequently the amplification of int2 gene has been used as a marker by some to study the amplification of other oncogenes at 1 l q l 3 , including cyclin Dl (Izzo et al., 1998; WmgetaL, 1999). However, other individuals have suggested that some of the alterations at 1 l q l 3 could involve deletions. These studies used microsatellite analysis and found A l in the 1 l q l 3 region, including int2, MEN1, 1 l q l 3 . 1 P F G M , DUS4946 and D11S913 loci (Bockmuhl et al, 1996; Hu i et al, 1996; D ' A d d a et al, 1999; Bikhazi et al, 2000; Iwasaki, 1996;Chakrabarti et al, 1998; Dwight et al, 2000; Guo et al, 2001; Nord et al, 1999). However, as discussed before, the limitation of microsatellite analysis is that it cannot distinguish deletion from amplification. Ah-See and his co-workers (1994) have shown that PCR-based assays cannot readily distinguish duplication or low-level amplification 39 of an allele from loss of heterozygosity, particularly i f there are contaminations of normal cells within the tumor. Nawroz et al. (1994) have stated that the L O H on 1 l q might in fact represent amplification at this region (1 l q l 3 ) . In contrast, in 1994, Ah-See et al used microsatellite analysis to score for A l and Southern blots to evaluate the amplification of CCND1 at the same time. The results demonstrated amplification in only one o f 20 tumors but A l in 9 of the 20 tumors. Such results would suggest that loss of TSGs at this region by deletion or mitotic recombination is a more l ikely scenario than amplification. In 1996, by using microsatellite assay and immunohistochemical analysis, Mura l i and coworkers found A l at 1 l q l 3 in 9 of 23 cancers, but amplification in only one of the 9 cancers, again pointing to a deletion theory. A l l these literature suggest that both amplification and deletion occur at 1 l q l 3 . One possible interpretation is that because TSGs (MEN1 etc) and oncogenes (CCND1, INT2, EMS1 etc) are both locate in this region, there might be different pathways of carcinogenesis, either involving amplification of oncogenes at l l q l 3 , or involving deletion of TSGs at l l q l 3 . N o matter whether the changes at 1 l q l 3 are deletion or amplification, the alterations have been correlated with clinical outcome, such as poor prognosis (Gebhart et al., 1998; Akerval l et al, 1997; Akerval l et al, 1995) and metastasis (Alavi et al, 1999). However, although many studies have investigated 1 l q in cancer, few have examined A l profiles in premalignant lesions, which is an open area for further research, and w i l l be a focus for this thesis. 40 Table 4. Amplification at l lql3 as identified with FISH in H N S C C Location Cases Reference l lql3 (CCND1) 13\/23 (56.5%) a Fujii et al, 2001 l lql3 8\/20 (40%) Ott et al, 2002 l lql3 (Zn*2) 12\/21 (57%) Bayerllein et al, 2000 l lql3 (CCND1) 5\/20 25%) A l a v i et al, 1999 l lql3 31\/85 (36%) Wil l iams etal, 1993 l lql3 146\/282 (52%) Muller etal, 1997 l lql3 22\/56 (39%) Meredith etal, 1995 l lq!3 11\/50(20%) Fortin etal, 1997 a Values given as number of samples showing loss\/total number o f informative cases. Values in parentheses are percentages. Mechanism of alteration at 1 lq22-23. In contrast, most of studies for the region of 1 lq22-23 indicate that alterations at 1 lq22-23 involve deletion. Besides studies using microsatellite analysis (Table 2 and Table 3), there are a few in which F I S H technique was used to study this region. Most of these F I S H studies were investigating leukemias. Only deletion was observed at the region spanning 1 lq22.3- l lq23 in some o f these studies (Leblanc et al, 1996; Dohner et al, 1997; Zhu et al, 1999; Lens et al, 2000; Zhu 41 et al, 2000; Cuneo et al, 2002). Interestingly, there are five articles that also reported amplification of Myelo id Lymphoid Leukemia gene (MLL gene) or 1 lq23, where the MLL gene is located, in leukemias by applying F I S H (Michaux et al, 2000; Cuthbert et al., 2000; Streubel et al, 2000; Reddy et al, 2001; Avet-Loiseau et al, 1999). In H N S C C , there is only one study applied F I S H to detect the alteration of 1 lq22-25 region. In this study, 5\/16 (31%) of adenomas and 2\/25 (8%) of primary hyperplasia were observed deleted at 1 lq23, but no amplification was observed. To date, no one has reported amplifications or oncogenes located at l lq22-25 involved in H N S C C . 42 1.3.4. l l q alterations in head and neck premalignant lesions Most of previous studies of 1 l q reported that alteration at 1 l q occurred late in carcinogenesis, between severe dysplasia and S C C (Table 2). So far, few papers have investigated l l q alterations in head and neck premalignant lesions. B y applying microsatellite analysis, el-Naggar et al. (1995), Califano et al. (1996) and Poh et al. (2001) have demonstrated A l at 1 l q in head & neck premalignant lesions (Table 5). In 2000, Rosin et al. (2000) indicated that L O H at 1 l q was strongly associated with an increase risk for progression. Their P-value was 0.062 and 0.011 for hyperplasia and low-grade dysplasia separately. A s we know, there are two hot spots at l l q , l l q l 3 and 1 lq22-23 regions, which play different roles in oral carcinogenesis. The limitation of this study is that it scored A l at these two sites together. This study needs to be confirmed using a larger number cases and scoring the two sites at 1 l q separately. Amplification at 1 l q l 3 has also been observed in head & neck premalignant lesions. In 1998, Izzo used F I S H to show that 7\/9 (77.7%) of dysplasia had amplification of CCND1 (Izzo et al, 1998). In one study, amplification of int2 gene (also done with FISH) was observed in 1\/4 (25%) during hyperplasia to dysplasia transition and 2\/4 (50%) of dysplasia (Roh et al, 2000). These two papers suggested that the int2-cyclinDl amplicon was amplified in the early stage in carcinogenesis of H N S C C and that cyclinDl might be involved in early regulation. However, the case numbers in both of these two papers are 43 very small and further study is required to confirm the results. These few papers demonstrate one or more of the following limitations: (1) they had limited number of cases; (2) they grouped all premalignant lesions together without separating them into different stages (e.g. different degree o f dysplasia) of premalignant lesions, and (3) they scored genetic alterations at 1 l q l 3 and 1 lq23 together (Table 5). It is still not well understood whether gene alterations at 1 l q precede carcinoma development or results from the unstable nature of tumors. Based on these limitations, more studies of genetic alterations at 1 l q in O P L need to be done to better understand the timing and mechanism of alterations of 1 l q in O P L and their clinical value. Table 5. Genetic alterations at l l q of head and neck premalignant lesions Analysis applied Location Stages Frequency Reference Microsatellite analysis l l q Premalignant lesions 9\/31 (29) a Califano et al, 1996 Microsatellite analysis H q Premalignant lesions 15\/75 (20) Rosin et al, 2000 Microsatellite analysis l l q Non-invasive lesion 1\/15 (7) el-Naggar et al, 1995 Microsatellite analysis l l q l 3 - 2 2 Hyperplasia 2\/44 (5) Poh et al, 2001 Low-grade dysplasia 6\/52 (12) High-grade dysplasia 6\/37 (16) a Values given as number of samples showing loss\/total number of informative cases. Values in parentheses are percentages. 44 1.3.5. Main genes on chromosome l l q Table 6 contains a list o f genes that are referred to in articles on 1 l q l 3 as possible oncogenes\/TSGs in that region. A t 1 lq23.1, the genes myeloid lymphoid leukemia gene (MLL gene), ALL-1, and Homology o f Trithoraz (HEX), are thought to play an important role in hematogenous malignancies with ATM being a key candidate for mutation in solid tumors, including H N S C C . A brief discussion of seven of these genes is given below. 45 Table 6. List of genes located at l l q l 3 GENE Location Function CCND1 Cyclin Dl 54917530-54930900 Regulating progress through the cell cycle FGF4\/HSTF1 Fibroblast growth factor 4 precursor 55049454-55051829 Heparin-binding growth factor FGF3\/int2 55086382-55087115 Stimulate the proliferation of fibroblast and endothelial cells FADD Fas associated via death domain 55347924-55351947 Universal adapter protein in apoptosis that mediates signaling of all known death domain-containing members of the T N F receptor superfamily EMS1 Cortactin 55543127-55581159 Involved in the restructuring of the cortical actin cytoskeleton RIN1 ras inhibitor 57291330-57295778 R A S inhibitor MEN1 Multiple endocrine neoplasia 1 59347865-59355064 Mutated form is responsible for Multiple endocrine neoplasia VEGFB\/VRF Vascular endothelial growth factor 59706252-59709238 Vascular endothelial growth factor 46 GENE Location Function PPP1CA\/PPP1 A 61059778-61063386 Serine\/threonine-specific protein phosphatases, A candidate T S G GST-pi Anionic glutathione S-transferase 61182590-61185105 Xenobiotic detoxification CPT1A Liver carnitine palmitoyltransferase I 62546480-6260940 Key enzyme in the carnitine-dependent transport across the mitochondrial inner membrane NUMA1 75601775-75679417 Cell-cycle related protein PAK1 p21-activated kinase 1 81163598-81233157 Regulate cytoskeletal dynamics by decreasing M L C K activity and myosin light-chain phosphorylation GARP Glycoprotein A repetitions predominate 81886288-81898610 A candidate oncogene at 1 lq l3 .5-l l q l 4 Location: From Human Genome project draft ( U C S C ) Dec. 22, 2001 browser (http:\/\/genome.cse.ucsc.edu\/index.html). 47 L3.5.a. CCND1 (cyclin Dl, PRAD1) sene at llgl3 The fidelity of cell division requires that an accurate copy of a complete genome be passed on to each daughter cell. This means that earlier events in the cell cycle, such as completion of D N A replication, must be accomplished for later events such as mitosis and cytokinesis to occur. Eukaryocytic cells have developed feedback controls called checkpoints to monitor and regulate various steps in cell-cycle progression. The CCND1 gene (cyclinDl gene) is a proto-oncogene that codes for a protein that is strongly implicated in cell cycle control. CCND1 binds to and activates a cell cycle kinase that controls phosphorylation of the retinoblastoma protein (pRb) (Donnellan et al., 1998). Phosphorylation to Rb gene causes it to release proteins such as transcription factor E2F. This factor can in turn bind to regulatory region in the D N A for specific genes and promote their transcription. This in turn drives cells through the checkpoint in late G l phase. Substantial evidence suggests that the level of cyclin D l protein is critical to proper cell cycle progression and that deregulated expression of this gene may disrupt cell cycle control and contribute to genomic instability (Almasan et al., 1995). Indeed, cells overexpressing cyclin D l have a shortened G l phase, reduced dependency on exogenous mitogens, and abnormal proliferative characteristics (Quelle et al, 1993; Jiang et al., 48 1993). These cells also demonstrate a higher frequency of gene amplification, especially under conditions of genotoxic stress (Zhou et al, 1996). Moreover, overexpression of cyclin D l has been shown to be associated with tumor transformation in in vitro studies in normal fibroblasts and primary embryo cells transfected with this gene (Jiang et al., 1993; Hinds et al., 1994; Uchimaru et al, 1996), and increased tumorigenesis in in vivo studies using transgenic mice that over-express this protein (Wang et al., 1994). CCND1 is commonly found amplified and overexpressed in almost all types of cancers, such as lung cancer (Reissmann et al., 1999), ovarian cancer (Dhar et al., 1999), myeloma (Hoechtlen-Vollmar et al., 2000), colorectal cancer (Mckay et al., 2000), prostate cancer (Kaltz-Wittmer et al., 2000), mantel-cell lymphoma (Remstein et al., 2000), breast cancer (Rennstam et al, 2001), and H N S C C (Julie et al, 1998). 1.3.5.b. int2 (FGF3) gene at Hal3 The gene int2 is also located in the chromosome 1 l q l 3 region. It is the first oncogene recognized as a member o f fibroblast growth factors (FGFs). Activation o f the int2 gene is a result of transcriptional deregulation, resulting in constitutive overexpression of the normal polypeptide products. The gene int2 encodes a growth factor known as F G F 3 , which is structurally related to other F G F family proteins and participates in several biological processes such as cell differentiation, motility, proliferation and angiogenesis (Dickson and Peters, 1987). A l l FGFs , including F G F 3 , have oncogenic potential. They 49 are synthesized by many tumor cells and can induce blood vessel formation (Burgess and Maciag, 1989; Goldfarb, 1990). Transferring int2 gene into mice results in hyperplasia of the mammary gland and prostate; however, tumor formation is rare (Muller et al, 1990; Ormtz etal, 1991). Genetic alteration of int2 has been reported mainly in H N S C C and breast cancer (Watatani et al, 2000; Selim et al, 2001, 2002; Lese et al, 1995; Rubin et al, 1995). A few papers also reported alteration olint2 in prostate carcinomas (Latil et. al, 1994) and ovarian cancer (Foulkes et al, 1993). In H N S C C , a few papers reported alteration at int2 as deletion (Friedman et al, 1989; Thakkar et al, 1989; Ah-See et al, 1994; Hu i et al, 1996). In contrast, alterations at int2 were recognized as amplification by more papers (Williams et al, 1993; Lese et al, 1995; Rubin et al, 1995; Mul le r et al, 1997; Jin et al, 1998; Wang et al, 1999; Shuster et al, 2000; Roh et al, 2000). I.3.5.C. HST1 sene at llql3 HST1 (Heparin secretory transforming factor 1, FGF4) is a human transforming gene originally detected by N I H 3T3 fibroblast transfection with D N A from human stomach tumors (Terada et al, 1986). It also belongs to the fibroblast growth factor family. Activation of HST1 is a result of transcriptional deregulation, resulting in constitutive overexpression of the normal polypeptide products. 50 In H N S C C , HST1 has been reported to be co-amplified with some genes at 1 l q l 3 (Sugimura et al, 1990; Lese et al, 1995; Meredith et al, 1995; Shuster et al, 2000). It is believed to be involved in H N S C C carcinogenesis (Lese et al, 1995) and angiogenesis (Schweigerer, 1989). I.3.5.d. EMS1 sene at llgl3 EMS1 [mammary tumor and squamous cell carcinoma-associated (p80\/85 src substrate)] gene has been mapped to 1 l q l 3 (Schuuring et al, 1992). It encodes human cortactin, an acting-binding protein possibly involved in the organization o f the cytoskeleton and cell adhesion structures (Shuster et al, 2000). Since amplification o f the 1 l q l 3 region has been associated with an enhanced invasive potential of these tumors (Adnane et al, 1989; Borg et al, 1991; Kitagawa et al, 1991; Schuuring et al, 1993), overexpression and concomitant accumulation of the EMS1 protein in the cell-substratum contact sites might contribute to the invasive potential of these tumor cells. EMS1 is generally believed to be an oncogene (Schuuring, 1995; Jin et al, 1998). Amplification oiEMSl gene has been observed in breast cancer (Schuuring et al, 1992; Brookes et al, 1993) and H N S C C (Williams et al, 1993; Jin et al, 1998; Shuster et al, 2000). Moreover, in H N S C C , 1 l q l 3 amplification, including EMS1, is associated with poor prognosis (Meredith et al, 1995; Rodrigo et al, 2000). 51 L3.5.e. MEN1 eene at llal3 Multiple endocrine neoplasia typel (MEN!) is an autosomal dominant disorder that is associated with endocrine tumors o f the parathyroid, the endocrine tissues, and the anterior pituitary (Wermer, 1954; Weber et al, \\99A). Additional associations include foregut carcinoid, facial angiofibroma, and lipomas. Larsson et al. (1998) initially made the critical observation that two malignant insulinomas from brothers with. MEN1 showed loss of the entire copy of chromosome l l q that was inherited from their parents without MEN1. This study suggested that the wi ld type MEN1 gene functions as a T S G . MEN1 was mapped to chromosome 1 l q l 3 in later studies (Nakamura et al, 1989; Bystrom et al, 1990; Janson et al, 1991). Other evidence supporting a role for MEN1 as a tumor suppressor gene comes from microsatellite studies that show L O H of the normal allele at the MEN1 locus (Vogelstein and Kinzler , 1998). Depending on the probes used, L O H has been shown to be frequent in MEN1 tumors of the parathyroid, approaching 100% (Lubensky et al, 1996; Zhuang et al, 1995). 1 l q l 3 L O H has also been found in 85% o f nongastrinoma pancreatic islet tumors and in 40% of gastrinomas (Debelenko et al, 1997). 1.3.5. f. RIN1 eene at 11a 13 Studies using F I S H on 10 oral S C C cell lines, which have been identified with 52 amplification of int2, HST1 and CCND1 shown that RIN1 (Ras interaction\/interference) gene is co-amplified (7 out of 10 cell lines) with other critical oncogenes, such as int2, CCND1 and HST1 in oral cancer (Shuster et al, 2000). Human RIN1 was first characterized as a R A S (an oncoprotein) binding protein based on the properties of its carboxyl-terminal domain (Han et al, 1997). Through a separate domain, RIN1 has also been shown to interact with and serve as a substrate for the tyrosine kinase A B L (an oncoprotein) (Afar et al, 1997)). The intimate relationship of RIN1 with both R A S and A B L oncoproteins suggests the possibility of a direct or indirect role for RIN1 in naturally occurring tumors (Shuster et al, 2000). I. 3.5.g. A TM eene at Hq23.1 ATM (Ataxia Telangiectasia Mutated) gene was found to be located on chromosome band 1 lq23.1 and its mutated form is responsible for ataxia telangiectasia (AT) . The A T M protein plays a key role in signaling cell cycle arrest in response to D N A double-strand breaks (Kastan et al, 1992; M e y n et al, 1994). Without this surveillance mechanism, cells are prone to replicate damaged D N A templates in S-phase and segregate damaged chromosomes through mitosis (Meyn et al, 1995). A T M protein also has been shown to interact with other proteins (Shafman et al, 1997) and to play a role in controlling cell cycle and apoptotic pathway (Barlow et al, 1997) as wel l as signal transduction (Keegan et al, 1996). Wild-type A T M protein is required for up-regulation 53 of p53 tumor suppressor protein in response to ionizing radiation and other D N A damaging agents (Westphal et al, 1997; Hawley et al, 1996). Cells lacking the A T M protein show a reduced and delayed activation o f the tumor suppressor gene p53 in response to D N A damage. The proposed function of the A T M protein points to a potential role of ATM as a tumor suppressor gene. A T is a hereditary autosomal recessive disorder with a variety of different clinical manifestations, including progressive cerebellar ataxia, oculocutaneous telangiectasis, immunodeficiency, chromosome instability, radiation sensitivity and an increased susceptibility for the development of various malignancies (Gatti et al, 1991). The incidence of the disease is estimated to be 1\/40,000-100,000 (Telatar et al, 1998; Gatti et al, 1991; Sedgwick et al, 1991). A T patients also exhibit severe hypoplasia of the thymus and lymphoid tissues, moderate to severe hypogonadism and atrophy of the cerebellum. A n estimated 1 in every 100 A T children from the age o f 10 onward w i l l develop a new cancer each year (Morrell et al, 1986; Taylor, 1992). The risk of developing cancer is 61 to 184 times higher in A T homozygotes than in the general population. More than 250 mutations in the v47Mgene have been identified in A T families (Vogelstein and Kinzler, 1998). L O H of ^ T M h a s been reported in breast cancer (Waha et al, 1998), cervical cancer (Skomedal et al, 1999), adult acute lymphoblastic leukemia (Haidar et al, 2000), ovarian cancer (Launonen et al, 1998), lung cancer (Murakami et al, 1999), and also H N S C C (Lazar et al, 1998). 54 Although these data suggest the possibility that the ATM gene could be acting as a T S G for these cancers, definitive proof is not available. For H N S C C , there have been no reports of mutation in this gene. 55 II. STATEMENT OF PROBLEMS II.l. Where are the additional tumor genes at l lql3 located? Although a number of tumor genes including CCND1 and int2 have been established in the 1 l q l 3 region, additional tumor genes, particularly tumor suppressor genes, are suspected to be located at the region. It has been proposed that there are many additional genes in the 1 l q l 3 region, and that this region may play a key role in controlling chromosomal instability and progression of tumors (Bekri et al, 1997, Izzo et al, 1998, 1999; Zhou et al, 1996; Gebhart et al, 1998). Identification and location of these potential genes w i l l contribute to the understanding of tumorigenesis. A major problem in using tumors to locate tumor genes is that cancers are most often accompanied by an intrinsic genetic instability that results in a cascade of gene alterations in the tumor, many of which are just random changes that are not critical to tumor development. This thesis used a spectrum of tissue samples including S C C , dysplasia and hyperplasia to search for novel regions of alterations in the 1 l q l 3 region. The rationale for using preinvasive lesions is that alterations occurring in these early lesions are more likely to be a driving force for carcinogenesis rather than due to the random genetic instability seen in later lesions including tumors. Consequently, studies of preinvasive lesions are more likely to reveal the small region of loss that contains critical tumor genes. 56 II.2. At what stages of oral cancer development does AI occur for int2 (llql3), D11S1778 (llq22.3) and any novel loci identified in this study? Information obtained from allelic imbalance studies has dual merit. The finding of frequently lost regions during cancer development can lead to discovery o f new TSGs. In addition, allelic imbalance findings can provide critical information on the role of the presumptive tumor genes even before the cloning of the tumor gene. For example, allelic imbalance studies have shown that there are three discrete regions of loss at 3p, suggesting that each of these three regions contains at least one tumor suppressor gene (Partridge et al., 1996). While we have yet to identify the genes involved at 3p, studies on the timing o f such loss during histological progression have already lead to the conclusion that loss on at least one of the regions, 3pl4 , is an early critical event for cancer development. A s discussed in section 1.2.3.f, L O H at 3 p l 4 can serve as an important molecular marker for predicting risk of malignant transformation for oral premalignant lesions (Rosin et al., 2000). While the mapping studies in this thesis may yield information on location of putative tumor suppressor genes at 1 lq l3 -14 , it may be some time before such genes can be identified. Acquisition of information on at what stage o f oral cancer development the new candidate gene is altered w i l l not only provide information on the possible roles of 57 the gene, but may also facilitate the process of identification of the gene. In addition to doing temporal studies for any novel loci showing frequent A l in the 1 l q l3 -14 region, this thesis w i l l further explore the timing of loss for the 2 regions previously studied in this lab (1.2.3.f). These regions are defined by markers int2 which amplifies sequence with the int2 gene at 1 l q l 3 and by marker D11S1778, which amplifies sequence that is 0.6 M b from the ATM gene at 1 lq22.3. Our early studies fused the data from these two loci (1.2.3.f). These studies suggested that alteration to 1 l q l 3 -22.3 occurred with formation of S C C . In this thesis, the 2 regions w i l l be examined independently to confirm the association with histological progression. 58 III. OBJECTIVES To use microsatellite analysis to examine D N A extracted from severe dysplasia, carcinoma in situ (CIS) and S C C for novel alterations in the 1 l q l 3 region. This region should be distinct from the int2 (1 l q l 3 ) and D11S1778 (1 lq22-23) regions of alteration previously studied in this laboratory. If a novel region of loss is identified, to determine at what stage of oral cancer development the alteration occurs by performing microsatellite analysis on a spectrum of stages o f oral premalignant lesions (hyperplasia, mi ld dysplasia, moderate dysplasia, severe dysplasia, CIS) as wel l as invasive S C C . To determine at what stage of oral cancer development the int2 (1 l q l 3 ) and D11S1778 (1 lq22.3) alteration occur by performing microsatellite analysis of the same oral premalignant and malignant lesions and compare the data obtained with that seen for any novel regions of loss identified in this study. To determine the significance of L O H at int2 (1 l q l 3 ) and D11S1778 (1 lq22.3) loci to the progression of oral premalignant lesions by comparing frequencies of loss for different locus in low-grade dysplasia with known outcome, i.e., low-grade lesions that did not progress into cancer with morphologically similar lesions that did develop into S C C . 59 IV. HYPOTHESES Microsatellite analysis of microdissected oral premalignant and malignant samples w i l l reveal novel loci in the 1 l q l 3 region, which harbors one or several putative tumor genes. Al le l i c loss at int2 (1 l q l 3 ) and D11S1778 (1 lq22-23) occur mainly at the tumor stage. Progressing low-grade lesions (those without dysplasia or with low-grade dysplasia) have increased frequencies of allelic imbalance at int2 (1 l q l 3 ) and D11S1778 (1 lq22-23) in this thesis compared to morphologically similar non-progressing lesions, suggesting a role for this alteration in cancer progression. 60 V. MATERIALS AND METHODS V . l . Sample collection This thesis used paraffin-embedded archival samples from the provincial Oral Biopsy Service of British Columbia, located at the Oral Pathology Divis ion of Vancouver General Hospital and Health Sciences Center. This service receives more than 3,500 biopsies of oral lesions received per year (19 years archived). This provides a large collection of early lesions that can be followed over time. The use of these samples was approved by the University Ethics Committee. V.2. Sample sets Two different sample sets were used. The first set had 4 groups in it, as follows: primary S C C , severe dysplasia\/CZS, mild\/moderate dysplasia and hyperplasia (Table 7). 61 Table 7. Histological groups in sample set 1 Lesion type Number of cases Hyperplasia (without dysplasia) 33 Mild & moderate dysplasia (low-grade dysplasia) 54 Severe dysplasia & CIS (high-grade dysplasia) 56 Primary SCC 91 Total 234 The second sample set included 2 groups of cases. Both groups included cases with hyperplasia (without dysplasia), mild dysplasia and moderate dysplasia (Table 8). In 1 group (called \"progressing lesions\") the lesions later progressed to CIS or S C C . No progression occurred in the second group (called \"non-progressing lesions\"). 62 Table 8. Histological groups in sample set 2: the progression test series Lesion Type Lesions not progressed to CIS or SCC Lesions later progressed to CIS or SCC Epithelial hyperplasia (without dysplasia) 33 6 Mild dysplasia 31 9 Moderate dysplasia 23 14 The criteria for choosing samples for the non-progressing group included confirmation of histological diagnosis by two pathologists using criteria established by the World Health Organization ( W H O collaborating Reference centre 1978) and the provision that the sample was large enough to yield sufficient D N A from both the epithelium and from the connective tissue for multiple microsatellite analyses. The third criterion was confirmation that these patients had no prior history of head and neck cancer and, for hyperplasia and dysplasia, that they did not subsequently develop such cancer. This confirmation was obtained from hospital records and by using a computer linkage with the British Columbia Cancer Registry. A l l but three o f these cases had at least 3 years of follow up. 63 The inclusion criteria for progressing group included confirmation of histological diagnosis by 2 pathologists, sufficient sample size, and no prior history o f H N S C C . A final provision was that both the primary hyperplastic or dysplastic lesions and their matching CIS or S C C had to be from the same anatomical site as recorded on pathology reports and patients charts and the interval between the primary lesions and later CIS or S C C had to be longer than 6 months. The later criterion was used to exclude cases where the appearance of the CIS or S C C might be due to inadequate biopsy or sampling o f the index (first) biopsy error. The time interval chosen for exclusion was arbitrary. V.3. Diagnostic criteria for the samples The W H O diagnostic were used, which have been reviewed in section I.2.2.b. The diagnosis was confirmed independently by Dr. R. Priddy and Dr. L . Zhang, oral pathologists at the B C provincial biopsy service. Only those cases in which the two pathologists agreed on the diagnosis were used for the study. V.4. Clinical information The following clinical data were obtained for the cases studied by examining pathology reports and hospital charts: smoking habit, age and gender of the patients and anatomical 64 location of the lesions. Some of this information was not recorded for some cases (see Result section). V.5. Slide preparation Tissue blocks for cases were chosen for study removed from the archive and one 5-micron-thick section was cut from each block, stained with H & E (hematoxylin and eosin) and coverslipped for reference. Further sections for microdissection were then cut at a 10 to 12 microns thickness with approximately 15 sections per sample. These sections were also stained with H & E but left uncoverslipped. The H & E procedure is described below: 1. Slides were baked at 37\u00b0C overnight in an oven, then at 60 to 65\u00b0C for 1 hour, and left at room temperature to cool. 2 Samples were deparaffmized by two changes o f xylene for 15 minutes. 3 Dehydration in gradient alcohols (100%, 95, 70% ethanol). 4 Hydration by rinsing in tap water. 5 Slides were placed in G i l l ' s Hematoxylin for 5 minutes then rinsed in tap water. 6. B l u e d \" with 1.5% (w\/v) sodium bicarbonate, then rinsed in water. 65 7. Slides were lightly counterstained with eosin, dehydrated, and cleared for coverslipping. 8. Thick sections to be dissected were stained by the above procedure without the dehydration step then air-dried prior to microdissection. V.6. Microdissection Microdissection of the specimens was either performed or supervised by Dr. L . Zhang. Areas o f hyperplasia, dysplasia and S C C were identified using the mounted H & E stained sections. Epithelial cells in chosen areas were meticulously microdissected from adjacent non-squamous epithelium tissue or cells under an inverted microscope using a 23G needle. Genomic D N A from normal tissue was obtained by dissecting out the underlying stroma in these sections. This D N A was used as control D N A for each case (Zhang al. 1997). V.7. Sample digestion and DNA extraction The microdissected tissue was collected in a 1.5 m l eppendorf tube and digested in 300 ul of 50 m M T r i s - H C L (pH 8.0) containing 1% sodium dodecyl sulfate (SDS) and proteinase K (0.5 mg\/ml) at 48\u00b0C for 72 or more hours. During incubation, samples were spiked with 10 or 20 p i of fresh proteinase K (20 mg\/ml) twice daily. The D N A was then 66 extracted 2 times with PC-9 , a phenol-chloroform mixture, precipitated with 100% ethanol in the presence o f glycogen, and washed with 70% ethanol. The samples were then re-suspended in L O T E , a low ionic strength Tris buffer, and submitted for D N A quantification (Rosin et al 1997; Zhang et al 1997). V.8. DNA quantification Fluorescence analysis with a Picogreen kit (Molecular Probes, Eugene, Oregon) was used to quantify D N A . This method used 2 standard curves. The low concentration standard curve was used for samples with 1 to 20 ng\/u.1, while the high concentration standard curve was used for concentrations between 10 and 400 ng\/ul. Absorbance was read with a S L M 4800C spectrofluorometer ( S L M Instruments Inc. Urbana, IL) . The sample D N A concentration was then determined from one of the standard curves depending on its concentration, hence absorbance. A series of dilutions were done subsequently to adjust the concentration of D N A to 5 ng\/ul with L O T E buffer (Rosin et al 1997; Zhang et al 1997). V.9. Primer extension preamplification (PEP) I f the concentration o f D N A was low (less than 100 ng total), a procedure called P E P was performed. P E P involves amplification of multiple sites of the genome using random 67 primers and low stringency conditions, thus increasing the amount o f total D N A for microsatellite analysis. The P E P reaction was carried out in a 60 p.1 reaction volume containing 20 ng of the D N A sample, 900 m M of T r i s - H C L , (pH 8.3), 2 m M of dNTP where N is A , C, G and T, 400 pJVI of random 15-mers (Operon Technologies, California), and 1 ul of Taq D N A polymerase (Introgen, Gibco). 2 drops of mineral o i l were added prior to the reaction. The amplification was done on an automated thermal cycler (Omigene H B T R 3 C M , Hybaid Ltd.) and involved 1 cycle o f pre-heat at 95\u00b0C for 2 minutes, followed by 50 cycles of: 1) denaturation at 92\u00b0C for 60 s, 2) annealing at 37\u00b0C for 2 min, and 3) polymerization at 55\u00b0C for 4 min (Rosin et al 1997; Zhang et al 1997). V.10. Coding samples A l l samples were coded in such a way that the analysis of allelic imbalance would be performed without the knowledge of the sample diagnosis. V . l l . End-Labeling One more step prior to microsatellite analysis was end-labeling o f one member of the chosen microsatellite primer pair. Those reactions were carried out in a total volume of 50 ul which contained 38 ul of P C R grade water, 5 ul of 10 x buffer for T4 polynucleotide kinase (New England BioLabs, Beverly, M A ) , 1.2 ul o f 100 x B S A , 100 68 ng of one of the primer pair, 3 ul of T4 polynucleotide kinase (New England BioLabs, Beverly, M A ) , and 2 ul of [y- 3 2P] A T P (20 uC i , Amersham). The P C R reaction was 1 cycle at 37\u00b0C for 60 min run on the thermal cycler (Rosin et al 1997; Zhang et al 1997) 69 V.12. Microsatellite analysis: PCR amplification The microsatellite markers came from Research Genetics (Huntsvile, A L ) and mapped to the following regions: 1 l q l 3 (D11S4207, D11S916 and int2) and 1 lq23 (D11S1778). Markers int2 was used to amplify D N A sequence within the int2 gene (need to confirm). Marker D11S1778 is 0.6 M b from the ATM gene and has been used as the marker for ^47Mgene in many publications (Laake K et al., 1997; 1999). P C R amplification was carried out in a 5 ul reaction volume containing 5 ng of genomic D N A , 1 ng of labeled primer, 10 ng o f each unlabeled primer, 1.5 m M each of d A T P , dGTP, dCTP, and dTTP, 0.5 units of Taq D N A polymerase (Invtogen, Gibco), P C R buffer [16.6 m M ammonium sulfate, 67 m M Tris (pH8.8), 6.7 m M magnesium chloride, 10 m M (P-mercaptoethanol, 6.7 m M E D T A , and 0.9% dimethyl sulfoxide], and 2 drops of mineral o i l . Amplification involved 1 cycle of pre-heat at 95C for 2 min; 40 cycles of 1) denaturation at 95 C\u00b0 for 30s, 2) annealing at 50-60 C\u00b0 (depending on the primer used) for 60s, and 3) polymerization at 70 C\u00b0 for 60s; and 1 cycle of final polymerization at 70 C\u00b0 for 5 min. The P C R products were then diluted 1:2 in loading buffer and separated on 7% urea-formamide-polyacrylamide gels, and visualized by autoradiography. The films were coded and scored for allelic imbalance (Zhang et al., 1997). 70 V.13. Scoring of allelic imbalance For informative cases (meaning both alleles are of different length and thus could be distinguished from one another by electrophoresis), allelic loss is scored i f the signal intensity of the band is at least 50% less than its normal control counterpart (connective tissue D N A ) (Rosin et al 1997; Zhang et al 1997). A l l samples showing allelic imbalance were subjected to repeat analysis after a second independent amplification and re-scored whenever the quantity of D N A is sufficient. 71 VI. RESULTS VI.l. Choice of microsatellite markers for this study There is strong evidence supporting the involvement of alterations in the 11 q l 3 region in ' cancer development and the potential involvement of numerous genes has been suggested (see Table 6 for list). However, with the exception of CCND1, the evidence for a role of these genes in oral carcinogenesis is largely speculative. When the research in this thesis was first started, a decision was made to focus on the D N A sequence between cyclin Dl (CCND1) and int2 in the 1 l q l 3 region, since many of these putative oncogene\/TSGs mapped to that region, and to use the microsatellite assay to screen for novel regions of alterations (distinct from CCND1). The rationale was that such novel alterations could be used to localize genes playing a role in oral carcinogenesis. The Human Genome Database (http:\/\/www.gdb.org\/) was used to identify potential microsatellite markers between CCND1 and int2. O f the markers chosen for the study, D11S4207 (AFMal03zf9) proved to be highly informative with frequent allelic loss in tumor D N A from microdissected oral S C C samples. It was thus chosen as a starting point for the study and the majority of the work described below used this marker. It should be noted that during the course of this research, there has been a vast improvement in sequence analysis for the genome. A s shown in Figure 6, this has resulted in the current localization of D i 1S4207 to a region that is 20 M b telomeric to the CCND1-Int2 region 72 [sequence localization by University o f California Santa Cruz database (http:\/\/genome.cse.ucsc.edu\/index.html) using the Human Genome project draft ( U C S C ) dated Dec. 22, 2001]. A s shown below, this marker identifies a novel region of frequent alteration in oral S C C and premalignant lesions that occurs independent of alterations at CCND1-Int2. The evidence in support of this statement is given below. 73 3 \u00ab o o CO a o Retenti CO CO Q o Retenti O \u00a9 # \u00a9 o C O \"EH C O u $ ^ cd \"I 2 rt C O cd \"EH C O >> CD *0 Cd r^5 o ^ <; cp u o c CU cu \u2022o a C3 s CO <\u2014i \u20224-1 cu co Cd cu u . O CU !. CU cu et CO O i -u s \u00a9 cu u a OJD =0 t\\ i ^ o m oo cu Co os 0\\ \u201e-< SO SO \u2022 x r - os \u2022 X co co -\u2014 c o -VO O o co ( X co co \u00a9 t \u00a7 \u00a9 \u00a9 @ @ \u00a9 \u00ae \u00a9 \u00a9 \u00a9 \u00a9 \u00ae \u00a9 S S S I I \u00a9 co o co SO OS co co \u2022 I X SO S oo oo Os i - i >X to I CO C M O i i CD 1 fi CO a fi CD > '5b fi - f i c3 O > PQ -a CD M CD _*o 'co fi O o co o o V O H CO u CD VI.4. Further evidence in support of the AI at D11S4207 being an independent event Another way of demonstrating that the A I at DUS4207 is occurring independent of alteration at the other 2 regions is to determine how frequently these alterations occur together or independent of each other in the same samples. Table 11 compares patterns of A I for D11S4207 and intl. Table 12 provides similar comparisons for D11S4207 and Dl 1 SI 778. A s shown in Table 11, in the majority of samples in which A I occurred at either D11S4207 or intl, the alteration in these 2 regions was not 'synchronous', that is, the alteration occurred in only 1 of the 2 primers and not both at the same time. This was true for 75% of hyperplasias, 100% of low-grade dysplasias73% of high-grade dysplasias, and 56% of SCCs . 85 Table 11. Patterns of alteration at D11S4207 and int2: frequencies at which these alterations occur together or independent of each other # of cases A l at A l at int2 A l at both Total # cases informative D11S4207 only D11S4207 with different at both loci only & int2 pattern (%)a Hyperplasia 13 2 1 1 3\/4 (75%) Low-grade dysplasia 22 5 1 0 6\/6 (100%) High-grade dysplasia 28 7 4 4 11\/15 (73%) S C C 51 9 9 14 18\/32 (56%) a Value given as number o f samples showing A l for one primer but retention for the other (% of cases in parentheses). Table 12 gives similar data for D11S4207 and D11S1778. Again, the alteration to these 2 regions was most often not synchronous. The alteration occurred in only 1 of the 2 regions in 100% of hyperplasias, 80% of low-grade dysplasias, 60% o f high-grade dysplasias and 70% of SCCs . 86 Table 12. Patterns of alteration at D11S4207 and D11S1778: frequencies at which these alterations occur together or independent of each other # of cases informative at both loci A la t D11S4207 only Ala t D11S1778 only A l at both D11S4207& D11S1778 Total # cases with different pattern (%)a Hyperplasia 15 5 0 0 5\/5 (100%) Low-grade dysplasia 29 8 0 2 8\/10 (80%) High-grade dysplasia 32 8 1 6 9\/15 (60%) SCC 63 18 8 11 26\/37 (70%) a Value given as number o f samples showing A l for one primer but retention for the other (%> of cases in parentheses). VI.5. Fine-mapping at D11S4207 The next logical step in this research was to fine-map the region around D11S4207 (using further primers) in order to determine the smallest region of A L That region would then be examined for putative oncogenes\/TSGs. A n attempt was made to do this early in the development of the thesis. Two primers were selected for study that originally mapped to 87 either side of D11S4207: D11S916 and D11S4119. The initial data suggested that we had successfully established telomeric and centromeric boundaries for this region of alteration. Unfortunately, a month ago, the location of these markers changed. The December 22,2001 Human Genome project draft shown in the University of California Santa Cruz database (http:\/\/genome.cse.ucsc.edu\/index.html) now places the markers as shown in Figure 6. Future studies w i l l require the careful mapping of primers onto tiled B A C arrays, a more stringent way of localizing markers. However, at present our data suggests that there may be a telomeric boundary at D11S4119 which is located approximately 0.5 megabase pairs from D11S4207. The data supporting this boundary is still weak with only 4 cases of S C C (161T, 414T, 448T and 529T) showing A l at D11S4207 and retention at D11S4119. VI.6. Allelic imbalance at chromosome l l q and malignant progression risk A n important question that must be answered with each marker that is proposed to be associated with progression is whether or not the presence of this alteration increases the risk of a premalignant lesion to transform into an invasive S C C . In order to answer this question, I compared A l frequencies in early premalignant lesions 88 with known outcome. Prior data on these samples has been presented in a recent published paper (Rosin et al. 2000) and is described below. A decision was made to look for associations of intl and D11S1778 to progression risk. D11S4207 was not studied because it requires further mapping to better localize the region o f alteration prior to using the primers on these very precious samples. The lesions consist of two groups of hyperplasia and oral low-grade dysplasia. One group (n - 87) is from patients with no subsequent history of head and neck cancer. A l l but three of these cases had at least 3 years of follow up time. We refer to these cases as 'non-progressing'. The other group (n = 29) is from patients that later progressed to CIS or S C C at the same anatomical site. The interval between the primary lesions and later CIS or S C C had to be longer than 6 months. We refer to them as 'progressing'. A s shown in Table 13, there was no significant difference between the progressing low-grade dysplasias and those without a history of progressing in terms of gender (56% male in progressing cases vs. 57% of those without a history), age (mean age 58 years in progressing cases versus 55 in those without a history), site, and smoking history (of those with known habits, 78% of progressing cases vs. 85% o f those without a history were smokers). However, on average, non-progressing cases were monitored for over twice the duration (96 versus 37 months) to ensure that progression did not occur. This lengthy follow-up time is to ensure that progression did not occur. 89 Table 13. Demographic information of patients with low-grade dysplasia Features Non-progressing (n=54) Progressing (n=23) P-value Age (mean, years) 55 58 0.416 Sex (% male) 57 56 1 % with smoking history 85 78 0.170 Follow-up (mean, months) 96 37 0.0001 Table 14 shows the association of A l at D11S1778 and int2 with progression. For non-progressing low-grade lesions, A l was rioted in only 4\/78 (5%) at D11S1778 and 4\/59 (7%) at int2. In contrast, A l was present 9\/23 (39%) at D11S1778 and 6\/21 (29%) at int2 in those morphologically similar but progressing lesions. The difference between progressing and non-progressing low-grade lesions is of statistical significance both at D11S1778 (p = 0.0002), and at int2 (p = 0.0175). 90 Table 14. Allelic imbalance of D11S1778 (ATM) and int2 (int2-cyclin Dl) in progressing and non-progressing hyperplasia and low-grade dysplasia Low-grade lesions Non-progressing (n=87) Progressing (n=29) p-value D11S1778 4\/78 (5 ) a 9\/23 (39) 0.0002b Int2 4\/59 (7) 6\/21 (29) 0.0175 a Value given as number o f samples showing loss\/total number of informative cases (% of cases in parentheses). b B o l d means p< 0.05, was considered significant. The data were also examined for association with disease progression by using the Kaplan-Meier method. Time-to-progression curves were plotted as a function of A I at D11S1778 or at int2 (figure 11). A significant difference was observed for each of the loci. These data suggest that D11S1778 and int2 both mark regions in the D N A with genes that are associated with risk of progression to cancer. 91 Figure 11. Probability of having no progression to cancer, according to A l at D11S1778 or at int2 VISITYR P = 0.0032 i. i 1.0' .9 co | M CO E 3 .7 1 I I I I I I -I\u2014I\u2014(-INT2 D 1.00 + 1.00-censored .00 4 .00-censored -10 0 VISITYR 10 20 P = 0.0054 92 VII. DISCUSSION To the best of my knowledge, this thesis has for the first time investigated a large number of premalignant lesions for genetic alterations at l l q . Three loci (Dl 1S4207, D11S1778 and intl) at l l q have been studied. D11S4207 is a hot locus newly identified by this thesis; whereas D11S1778 and int2 are two loci that are widely studied in cancer but only limited information is available regarding their occurrences in oral premalignant lesions. For a better flow o f the discussion, I w i l l first discuss the temporal changes o f the 3 loci during the multistage oral carcinogenesis, and then discuss the relationship between cancer progression of oral premalignant lesions and l l q changes, and finally the new locus, D11S4207, as a potential new hot spot containing tumor gene(s). VII.l. Allelic imbalance of genes at D11S4207, D11S1778 and int2-cycline Dl during multistage oral carcinogenesis The histological progression model for head and neck S C C is wel l established. There are strong evidences indicating that accumulations of changes to critical control genes (oncogenes and TSGs) underline the progression of lesions from hyperplasia to increasing degree o f dysplasia (mild, moderate, severe), and to CIS and finally to invasive S C C . Currently a number of tumor genes and even more potential chromosome loci containing tumor genes have been established in oral S C C but information on these genes 93 and loci in the early stage of oral lesions are limited in number and scope due to the difficulty of obtaining suitable specimens for analysis and to technical problems associated with working with very small lesions and minute amounts o f D N A (Zhang et al, 1997; Califano et al, 1996; Roz et al, 1996; Mao et al, 1996b; Emil ion et al, 1996). Although a molecular progression model has been first proposed by Califano et al. (1996), and later refined by Rosin et al. (2000), genes and chromosome loci investigated in the model are limited. Understanding of the additional genetic changes and of their timing during the molecular progression of oral cancer is critical for our further understanding of the mechanisms of the tumor progression and prediction of cancer risk o f oral premalignant lesions as well as intervention and management o f high-risk oral lesions. This thesis has studied a large number of oral premalignant lesions at different stages of progression for alterations for chromosome 1 l q at D11S4207, D11S1778 (1 lq22-23) and intl (int2-cyclin Dl region) to determine the timing of the alterations at these loci . VII.1.1. Temporal changes of the 3 loci at l l q This thesis provides a great deal of genetic information, for the first time, of timing and frequency of A l at these three loci in oral carcinogenesis, which is essential to help better understanding of the mechanism of oral carcinogenesis and the role of 1 l q . 94 Temporally, allelic imbalance at D11S4207 (the new gene site) is markedly different from those at D11S1778 (1 lq22-23) and intl (int2-cyclin Dl region). A I for D11S4207 (the new gene site) occurred very early during the multistage carcinogenesis with only slight increase in the frequency of A I with progression of the oral lesions from low-grade to high-grade lesions and finally to invasive S C C . In oral hyperplasia, 7\/19 (37%) demonstrate allelic imbalance for the new gene site. This rate is much higher than that seen for 3p and 9p losses in oral hyperplasia, both of which have been shown to occur early in the development o f oral and many other solid cancers (Rosin et al, 2000). However, unlike the changes seen for 3p and 9p losses, which steadily increase with progression of oral lesions, the rate o f allelic imbalance seen in oral hyperplasia at D11S4207 (the new gene site) has not markedly increased with progression o f the oral lesions. Even when the rate of the allelic imbalance of the invasive oral S C C (35\/74, 47%) is compared with that of the oral hyperplasia (7\/19. 37%), the result is far from significantly different (P = 0.4506). Unlike the new gene locus, allelic imbalances at Dl 1 SI 778 ( l lq22-23) and int2 (int2-cyclin Dl region) are rare at low-risk oral premalignant lesions. Both o f these markers show markedly increased allelic imbalance with advent of high-grade dysplasias, which contain significantly increased alterations at these markers compared to low-grade dysplasia and hyperplasia (P = 0.0352 for D11S1778 and P = 0.0223 for intl). The rise in the frequency of allelic imbalance continued in the invasive SCCs . 95 VII. 1.2. Significance of the changes of the 3 loci at 11 q Studies of temporal changes o f allelic imbalance during multistage oral carcinogenesis can provide critical information on the role of a presumptive tumor gene in the development o f oral cancer even prior to the identification o f the actual tumor gene. For example, i f a gene were mainly altered during the transformation of preinvasive lesions to invasive lesions, it would suggest that the gene plays a role in the tumor invasion. Vn.1.2.1. Sienificance of Al at D11S4207 The finding of a high frequency of A l at D11S4207 (the new gene site) in oral lesions with very low cancer risk (37% of hyperplasia) and the finding of a similar frequency or only slightly higher frequency of A l in oral lesions with higher cancer risk (34.5% low-grade dysplasia and 47% of high-grade dysplasia) or even in S C C (47%) are unusual. Our previous studies on other chromosome regions in oral lesions generally demonstrate a significant rise in the losses at a specific stage(s) of oral cancer development. For example, losses at 3p and 9p were found to be significantly (or approaching significantly) increased from hyperplasia to low-grade dysplasia, and then continued to be markedly increased at later stages (high-grade dysplasia or S C C ) . On the other hand, a significant increase in L O H at 17p occurred with the formation of high-grade dysplasias, whereas significant increases in L O H at 8p and 13q occurred with the advent of invasive S C C (see Table 1 at Section 1.2.3.f). 96 Is such early occurrence of a high frequency of A l at D11S4207 without obvious increase in the frequency of A l with progression of oral lesions exceptional? In a recent study, we have noticed similar temporal changes o f A l in another chromosome region with presumptive tumor genes during oral carcinogenesis. A l at 14q31-32 was shown to be high in low-grade dysplasia (17\/51, 33%), but the frequency o f A l did not increase with progression of oral lesions: A l was noted in 10\/23 (30%) of high-grade dysplasias and 12\/33 (36%>) invasive SCCs (unpublished data from this lab). The significance o f such early occurrence of A l without accompanying increase with progression o f oral lesions remains speculative. Since demonstration o f A l by microsatellite analysis is an evidence of clonal expansion of cells with growth advantages (required by tumorigenesis), the presence of A l , even in hyperplasia, could not be ignored. The explanation for lack of further increase in the frequency o f A l could be complicated, including alternate modes of inactivation\/alteration such as epigenetic silencing of gene expression by promoter methylation (Thiagalingam et al., 2002), and different cancer development pathways, or simply because only one primer, D11S4207, has been used to probe the region, and an increased number o f primers for the region may identify further cases with the alteration. Whatever function or significance the new locus has, it is clear that A l at D11S4207 occurred independently of A l at int2 (D11S4207 is localized at a region 20 M b telomeric to CCNDl-int2 region), and of A l at D11S1778 (1 lq22-23). The results in Section VI.4 97 c1 \\ showed that the majority of samples did not show A I at D11S4207 simultaneously or synchronously with either A I at int2 or at D11S1778 (see Tables 11 and 12). A t int2, the discordance with D11S4207 occurred in 50% of hyperplasias, 100% of low-grade dysplasias, 73% o f high-grade dysplasias, and 56% of SCCs . A t D11S1778, the discordance D11S4207 occurred in 100%) of hyperplasias, 80% of low-grade dysplasias, 60%> of high-grade dysplasias and 70% of SCCs . The rates should be even higher given the fact that the assay does not yield any information on whether or not the maternal or paternal allele is being altered in samples showing A I . Thus, estimates on the actual percentage of cases in which 2 events are giving rise to alterations at these 2 loci are even lower. A n alteration could occur on the maternal allele of one locus and on the paternal allele of another loci and still be recorded as a single event. Several studies have suggested the existence o f fragile sites at 1 l q l 3 region (Conquelle et al, 1997; Jin et al, 1998; Shuster et al, 2000), including the region harboring D11S4207 (Jin et al, 1998). One speculation for the early occurrence of A I at D11S4207 is that cells with such fragile site alterations have an increased genomic instability, which subjects the cells with the A I to further genetic changes. In our recent publication, it has been shown that A I at 3p &\/or 9p occurred early and were found in almost all oral premalignant lesions that later progressed into cancer, suggesting A I at these loci are essential for cancer formation. Table 15 examines the relationship between A I at D11S4207 and A I at 3p &\/or 9p, the proposed essential changes for oral cancer development, to determine the potential importance of A I at D11S4207. 98 O f the 29 cases of low-grade dysplasia, 10 cases have A l at D11S4207, whereas 19 have no alteration at the site. A l at 3p &\/or 9p was noted in a significantly higher proportion of lesions with A l at D11S4207 as compared to lesions without A l at D11S4207: 9 (90%) of 10 lesions vs. 7 (37%) of 19 cases (P = 0.0084). Such results suggest an association between genetic alterations at D11S4207 and those at 3p and 9p. Table 15. Association of allelic imbalance at D11S4207 and 3p &\/or 9p in low-grade dysplasias 3p and\/or 9p P-value A l Retention D11S4207 A l 9 1 0.0084 Retention 7 12 VII.1.2.2. Significance ofAIatDHS1778 (11Q22-23) and int2 (int2-cyclin Dl region) Although A l at 1 lq l3 -22 has been extensively investigated in head and neck cancer, this is the first study to investigate int2-cyclin Dl genes at 11 q 13 and 1 lq22-23 region separately in oral premalignant lesions with a large number of oral lesions at different stages o f cancer progression. A l at the two loci was rare in hyperplasia and low-grade dysplasia. Both D11S1778 (1 lq22-23) and int2 (int2-cyclin Dl region) showed significant increase in the frequency of A l with formation of high-grade oral 99 dysplastic lesions, and the increase continued with formation o f S C C . Such results indicate that lesions with these changes have high-risk for cancer progression, and also support the literature that genes at 1 lq l3 -22 play important roles in tumor instability and invasion. VII.2. Allelic imbalance at D11S1778 and int2 is associated with cancer risk. A salient advantage for microsatellite analysis is that data from microsatellite analysis not only could provide clue for the identification o f target genes, but also could be used as markers for diagnosis and prognosis, even prior to identification o f the actual target genes. The latter is particularly important as from a clinical point o f view; the ultimate significance of molecular studies is that molecular markers can be used to guide clinical management, including prediction of risk of cancer progression of oral premalignant lesions. Despite o f the obvious significance and importance o f l inking molecular results with clinical outcome, few such studies are available for premalignant lesions. Genetic studies of oral premalignant lesions are difficult even without the requirement of the clinical outcome. In this study, microdissected early oral premalignant lesions from 116 patients with or without a history of progression into CIS or invasive S C C were analyzed for A I at D11S1778 and int2 in order to determine the potential roles of these gene loci in the 100 progression of low-grade oral premalignant lesions. A s discussed above, the results of one of my studies show that allelic imbalances at D11S1778 and int2 occur mainly in high-grade oral dysplastic lesions and invasive oral SCCs . These results would suggest that these molecular markers could serve as risk markers for cancer progression since high-grade dysplasias are known to have a high-risk for cancer progression. Low-grade lesions (those without dysplasia or with low-grade dysplasia) were chosen for this study because the majority of these lesions w i l l not progress into cancer. Currently it is not possible to identify the small percentage of progressing low-grade lesions from the majority of morphologically similar but non-progressing lesions. Since A l at D11S1778 and int2 were rare in the low-grade lesions in my study, in this study we asked whether A l at D11S1778 and int2 indicate a risk of cancer progression, and whether those low-grade dysplasias with A l at the two loci had increased cancer risk. The study showed that non-progressing low-grade lesions (without dysplasia or with low-grade dysplasia) had significantly lower frequencies of allelic imbalance at both D11S1778 and int2, as compared to the non-progressing hyperplasia (Table 14). For non-progressing low-grade lesions, A l was noted in only 4\/78 (5%) at D11S1778 and 4\/59 (7%) at int2, significantly lower than those morphologically similar but progressing lesions: 9\/23 (39%) at D11S1778 (P = 0.0002) and 6\/21 (29%) at int2 (P = 0.0175). Such study results lend strong support to the hypothesis that A l at both D11S1778 101 (1 lq22-23) and int2 (int2-cyclin Dl region) is associated with high cancer risk for oral premalignant lesions and that morphologically low-grade lesion with A I at either of the two loci may indicate high cancer risk despite of a low-risk morphology. This thesis provide solid evidence that A I at D11S1778 and int2 could be used as potential markers to identify high-risk lesions at early stage, which may have important impact on the clinical diagnosis and management. VI1.3.D11S4207, a new hot spot at llq!3 A s mentioned in literature review, despite o f discovery o f a number of genes at the 1 l q l 3 region, it is generally believed that many more have yet to be identified from the site, which is a major region that plays a key role in controlling chromosomal instability and progression of tumors (Bekri et al, 1997, Izzo et al, 1998, 1999; Zhou et al, 1996; Gebhart et al, 1998). Identification and localization of these potential genes w i l l contribute to the understanding of the roles of 1 l q l 3 region in tumorigenesis, including oral carcinogenesis. This thesis has identified a new hot spot, D11S4207, for tumor gene at 1 l q l 3 region. O f the 91 oral S C C investigated, 74 were informative and 35\/74 (47%) demonstrated allelic imbalance at Dl 1S4207. Subsequent to the discovery of the new locus, I investigated the boundaries of the region 102 of alteration. M y data had indicated successful establishment of telomeric and centromeric boundaries for the region of alteration with microsatellite markers D11S916 and D11S4119, until one month ago when the location of these markers changed (http:\/\/gonome.cse.ucsc.edu\/index.html), although two cases of S C C did suggest that there may be a telomeric boundary at D11S4119, which is located approximately 0.5 megabase pairs from D11S4207. Despite of the need for further fine mapping of the region, existing data do support that this is a new hot spot containing tumor gene(s). Not only there is a high frequency of A l at D11S4207 in oral SCCs , a high frequency of A l at D11S4207 was also noted in preinvasive lesions including low risk oral hyperplasias. Cancers are characterized by an intrinsic genetic instability that frequently results in a cascade o f nonspecific genetic alterations, which make the identification of alterations to critical control genes difficult. Consequently it could be argued that A l at D11S4207 in oral S C C s could be non-specific, even though the frequency of A l (47%) is too high for that argument. The presence of A l at the same region in preinvasive lesions, however, strongly support that the alteration is not random 'gun-shot' effects, but rather true hot spot containing tumor gene(s). The identification of this new hot spot and further fine mapping o f the region w i l l lead to pinpoint the location of putative tumor genes for further analysis and is critical for further sequencing to identify new genes and related functions, and add further understanding to the role of 1 l q l 3 in the cancer development. 103 VIIAEnding Mark Although my study identified a new locus and also provides A I frequencies of three loci at 1 l q : D11S4207, int2 and D11S1778 in oral cancer and premalignant lesions, there are some limitations as well . One major limitation is that there was only one primer used for each locus of the three loci at 1 l q . This is insufficient since each o f the loci probably contains multiple genes; hence the data provide only information of 'global ' changes on the region. The use of only one primer for each locus also increases the percentages of cases studied being non-informative, which lowers the sample size with data and increases the possibility o f Type II error in statistical analysis. For example, 14 (42%) out of 33 of hyperplasia cases were non-informative at D11S4207. There are many exciting things I would like to continue on the project in the future. This w i l l include fine map the locus of D11S4207 to localize and ultimately sequence and identify the candidate tumor genes in this region. Currently it seems that careful mapping o f primers onto tiled B A C arrays would be the choice o f techniques although other newer techniques could be out very soon with the rapid advances in technologies. The identification of the gene(s) could be followed by immunohistochemical studies of the gene products to investigate the protein changes of the genes. I would also like to employ other techniques, such as F I S H to determine whether the alteration at the new locus is deletion or amplification. The information w i l l be important for determination whether the genes are tumor suppressor genes (deletion) or oncogenes (amplification) and further mechanistic studies. 104 BIBIOGRAPHY 1. Adnane J, Gaudray P, Simon M P , et al. Proto-oncogene amplification and human breast tumor phenotype. Oncogene 1989; 4(l l) :1389-95. 2. Afar D E , Han L , McLaughl in J, et al. Regulation of the oncogenic activity of B C R - A B L by a tightly bound substrate protein R I N 1 . Immunity 1997; 6(6):773-82. 3. Ah-See K W , Cooke T G , Pickford IR, et al. A n allelotype o f squamous carcinoma of the head and neck using microsatellite markers. Cancer Res 1994; 54(7):1617-21. 4. Akerval l J A , Jin Y , Wennerberg JP, et al. Chromosomal abnormalities involving 1 l q l 3 are associated with poor prognosis in patients with squamous cell carcinoma of the head and neck. Cancer 1995; 76(5):853-9. 5. Akerval l J A , Michalides R J , Mineta H , et al. Amplification of cyclin D l in squamous cell carcinoma of the head and neck and the prognostic value of-chromosomal abnormalities and cyclin D l overexpression. Cancer 1997; 79(2):380-9. 105 6. A l a v i S, Namazie A , Calcaterra T C , et al. Cl inical application of fluorescence in situ hybridization for chromosome 1 l q l 3 analysis in head and neck cancer. Laryngoscope 1999; 109(6):874-9. 7. Almasan A , Linke SP, Paulson T G , et al. Genetic instability as a consequence of inappropriate entry into and progression through S-phase. Cancer Metastasis Rev 1995; 14(l):59-73. 8. Anderson J A , Irish JC, McLach l in C M , Ngan B Y . H-ras oncogene mutation and human papillomavirus infection in oral carcinomas. A r c h Otolaryngol Head Neck Surg 1994; 120(7):755-60. 9. Avet-Loiseau H , Godon C, L i J Y , et al. Amplification o f the 1 lq23 region in acute myeloid leukemia. Genes Chromosomes Cancer 1999; 26(2): 166-70. 10. A x e l l T, Pindborg JJ, Smith CJ , van der Waal I. Oral white lesions with special reference to precancerous and tobacco- related lesions: conclusions of an international symposium held in Uppsala, Sweden, M a y 18-21 1994. International Collaborative Group on Oral White Lesions. J Oral Pathol M e d 1996;25(2):49-54. 11. Banoczy J, Csiba A . Occurrence of epithelial dysplasia in oral leukoplakia. Analysis and follow-up study of 12 cases. Oral Surg Oral M e d Oral Pathol 106 1976; 42(6):766-74. 12. Barghorn A , Komminoth P, Bachmann D , et al. Deletion at 3p25.3-p23 is frequently encountered in endocrine pancreatic tumours and is associated with metastatic progression. J Pathol 2001; 194(4):451-8. 13. Barlow C, Brown K D , Deng C X , et al. A t m selectively regulates distinct p53-dependent cell-cycle checkpoint and apoptotic pathways. Nat Genet 1997; 17(4):453-6. 14. Bartkova J, Lukas J, Mul ler H , et al. Abnormal patterns o f D-type cyclin expression and G l regulation in human head and neck cancer. Cancer Res 1995; 55(4):949-56. 15. Bartkova J, Lukas J, Strauss M , Bartek J. Cyc l in D l oncoprotein aberrantly accumulates in malignancies of diverse histogenesis. Oncogene 1995; 10(4):775-8. 16. Baskaran R, Wood L D , Whitaker L L , et al. Ataxia telangiectasia mutant protein activates c -Ab l tyrosine kinase in response to ionizing radiation. Nature 1997; 387(6632):516-9. 17. Bayerlein K , Rith T, Verdorfer I, et al. I -FISH control o f CGH-detected gain 107 of D N A sequence copy number in oral squamous cell carcinomas (OSCC) . Anticancer Res 2000; 20(1 A):427-32. 18. Beckman JS, Weber JL. Survey of human and rat microsatellites. Genomics 1992; 12(4):627-31. 19. Bikhazi P H , Messina L , Mhatre A N , et al. Molecular pathogenesis in sporadic head and neck paraganglioma. Laryngoscope 2000; 110(8):1346-8. 20. Bockmuhl U , Petersen I, Dietel M . [Allele losses in squamous cell carcinomas of the larynx]. Laryngorhinootologie 1996; 75(l):48-62. 21. Bockmuhl U , Schwendel A , Dietel M , Petersen I. Distinct patterns of chromosomal alterations in high- and low-grade head and neck squamous cell carcinomas. Cancer Res 1996; 56(23):5325-9. 22. Bockmuhl U , Petersen I, Schwendel A , Dietel M . [Genetic screening of head-neck carcinomas using comparative genomic hybridization (CGH)] . Laryngorhinootologie 1996; 75(7):408-14. 23. Boone C W , Kel lof f GJ , Steele V E . The natural history o f intraepithelial neoplasia: relevance to the search for intermediate endpoint biomarkers. J Cel l Biochem Suppl 1992:23-6. 108 24. Borg A , Sigurdsson H , Clark G M , et al. Association o f INT2\/HST1 coamplification in primary breast cancer with hormone-dependent phenotype and poor prognosis. B r J Cancer 1991; 63(l):136-42. 25. Bouquot JE, Ephros H . Erythroplakia: the dangerous red mucosa. Pract Periodontics Aesthet Dent 1995; 7(6):59-67; quiz 68. 26. Braithwaite K L , Rabbitts P H . Multi-step evolution of lung cancer. Semin Cancer B i o l 1999; 9(4):255-65. 27. Brookes S, Lammie G A , Schuuring E , et al. Amplif ied region of chromosome band 1 l q l 3 in breast and squamous cell carcinomas encompasses three C p G islands telomeric of F G F 3 , including the expressed gene E M S 1 . Genes Chromosomes Cancer 1993; 6(4):222-31. 28. Burgess W H , Maciag T. The heparin-binding (fibroblast) growth factor family of proteins. Annu Rev Biochem 1989; 58:575-606. 29. Bystrom C, Larsson C, Blomberg C, et al. Localization of the M E N 1 gene to a small region within chromosome 1 l q l 3 by deletion mapping in tumors. Proc Natl Acad Sci U S A 1990; 87(5): 1968-72. 30. Cairns P, Sidransky D . Molecular methods for the diagnosis o f cancer. 109 Biochim Biophys Acta 1999; 1423(2):C11-8. 31. Califano J, van der Riet P, Westra W , et al. Genetic progression model for head and neck cancer: implications for field cancerization. Cancer Res 1996; 56(ll):2488-92. 32. Carter S L , Negrini M , Baffa R, et al. Loss of heterozygosity at 1 Iq22-q23 in breast cancer. Cancer Res 1994; 54(23):6270-4. 33. Chakrabarti R, Srivatsan E S , Wood TF , et al. Deletion mapping of endocrine tumors localizes a second tumor suppressor gene on chromosome band 1 l q l 3 . Genes Chromosomes Cancer 1998; 22(2): 130-7. 34. Clark L J , Edington K , Swan IR, et al. The absence o f Harvey ras mutations during development and progression of squamous cell carcinomas of the head and neck. B r J Cancer 1993; 68(3):617-20. 35. Cooper E H , de Mel lo Junior JP, Giles G R . Biochemical markers in gastrointestinal malignancies. A r q Gastroenterol 1989; 26(4):131-40. 36. Coquelle A , Pipiras E , Toledo F, et al. Expression of fragile sites triggers intrachromosomal mammalian gene amplification and sets boundaries to early amplicons. Ce l l 1997; 89(2):215-25. 110 37. Croce C M , Sozzi G , Huebner K . Role of FHIT in human cancer. J C l i n Oncol 1999; 17(5):1618-24. 38. Cuneo A , Roberti M G , Bigoni R, et al. Four novel non-random chromosome rearrangements in B-cel l chronic lymphocytic leukaemia: 6p24-25 and 12pl2-13 translocations, 4q21 anomalies and monosomy 21. B r J Haematol 2000; 108(3):559-64. 39. Cuthbert G , Thompson K , McCul lough S, et al. M L L amplification in acute leukaemia: a United Kingdom Cancer Cytogenetics Group ( U K C C G ) study. Leukemia 2000; 14(11):1885-91. 40. D 'Adda T, Keller G , Bordi C , Hofler H . Loss o f heterozygosity in 1 l q l 3-14 regions in gastric neuroendocrine tumors not associated with multiple endocrine neoplasia type 1 syndrome. Lab Invest 1999; 79(6):671-7. 41. Day G L , Blot W J , Shore R E , et al. Second cancers following oral and pharyngeal cancers: role of tobacco and alcohol. J Natl Cancer Inst 1994; 86(2): 131-7. 42. Debelenko L V , Zhuang Z , Emmert-Buck M R , et al. A l l e l i c deletions on chromosome 1 l q l 3 in multiple endocrine neoplasia type 1-associated and sporadic gastrinomas and pancreatic endocrine tumors. Cancer Res 1997; 111 57(ll):2238-43. 43. Dhar K K , Branigan K , Parkes J, et al. Expression and subcellular localization o f cyclin D l protein in epithelial ovarian tumour cells. B r J Cancer 1999; 81(7):1174-81. 44. Dohner H , Stilgenbauer S, James M R , et al. l l q deletions identify a new subset of B-cel l chronic lymphocytic leukemia characterized by extensive, nodal involvement and inferior prognosis. Blood 1997; 89(7):2516-22. 45. Donnellan R, Chetty R. Cycl in D l and human neoplasia. M o l Pathol 1998; 51( l ) : l -7 . 46. Dwight T, Twigg S, Delbridge L , et al. Loss of heterozygosity in sporadic parathyroid tumours: involvement o f chromosome 1 and the M E N 1 gene locus in l l q l 3 . C l i n Endocrinol (Oxf) 2000; 53(l):85-92. 47. el-Naggar A K , Steck K , Batsakis JG . Heterogeneity o f the proliferative fraction and cyclin D 1 \/ C C N D 1 gene amplification in head and neck squamous cell carcinoma. Cytometry 1995; 21(1):47-51. 48. el-Naggar A K , Hurr K , Batsakis J G , et al. Sequential loss of heterozygosity at microsatellite motifs in preinvasive and invasive head and neck squamous 112 carcinoma. Cancer Res 1995; 55(12):2656-9. 49. Fearon E R , Vogelstein B . A genetic model for colorectal tumorigenesis. Cel l 1990; 61(5):759-67. 50. Fearon E R . Human cancer syndromes: clues to the origin and nature of cancer. Science 1997; 278(5340): 1043-50. 51. Fortin A , Guerry M , Guerry R, et al. Chromosome 1 l q l 3 gene amplifications in oral and oropharyngeal carcinomas: no correlation with subclinical lymph node invasion and disease recurrence. C l i n Cancer Res 1997; 3(9):1609-14. 52. Foulkes W D , Campbell IG, Stamp G W , Trowsdale J. Loss of heterozygosity and amplification on chromosome 1 l q in human ovarian cancer. B r J Cancer . 1993; 67(2):268-73. 53. Friedman E , Sakaguchi K , Bale A E , et al. Clonality o f parathyroid tumors in familial multiple endocrine neoplasia type 1. N Engl J M e d 1989; 321(4):213-8. 54. Fujii M , Ishiguro R, Yamashita T, Tashiro M . Cyc l in D l amplification correlates with early recurrence of squamous cell carcinoma o f the tongue. Cancer Lett 2001; 172(2): 187-92. 113 55. Gabra H , Taylor L , Cohen B B , et al. Chromosome 11 allele imbalance and clinicopathological correlates in ovarian tumours. B r J Cancer 1995; 72(2):367-75. 56. Gabra H , Watson JE, Taylor K J , et al. Definition and refinement of a region of loss of heterozygosity at 1 Iq23.3-q24.3 in epithelial ovarian cancer associated with poor prognosis. Cancer Res 1996; 56(5):950-4. 57. Gallo O, Chiarelli I, Boddi V , et al. Cumulative prognostic value of p53 mutations and bcl-2 protein expression in head-and-neck cancer treated by radiotherapy. Int J Cancer 1999; 84(6):573-9. 58. Gatti R A , Boder E , Vinters H V , et al. Ataxia-telangiectasia: an interdisciplinary approach to pathogenesis. Medicine (Baltimore) 1991; 70(2):99-117. 59. Gebhart E , Liehr T, W o l f f E , et al. Pattern o f genomic imbalances in oral squamous cell carcinomas with and without an increased copy number of l l q l 3 . Int J Oncol 1998; 12(5):1151-5. 60. Geboes K . Barrett's esophagus: the metaplasia-dysplasia-carcinoma sequence: morphological aspects. Acta Gastroenterol Be lg 2000; 63(l):13-7. 114 61. Gentile M , Olsen K , Dufmats M , Wingren S. Frequent allelic losses at 11 q24.1 -q25 in young women with breast cancer: association with poor survival. B r J Cancer 1999; 80(5-6):843-9. 62. Gleich L L , L i Y Q , Biddinger P W , et al. The loss o f heterozygosity in retinoblastoma and p53 suppressor genes as a prognostic indicator for head and neck cancer. Laryngoscope 1996; 106(11): 1378-81. 63. Goldfarb M . The fibroblast growth factor family. Ce l l Growth Differ 1990; l(9):439-45. 64. Goodarzi G , Mashimo T, Watabe M , et al. Identification of tumor metastasis suppressor region on the short arm of human chromosome 20. Genes Chromosomes Cancer 2001; 32(l):33-42. 65. Greenlee R T , Hil l -Harmon M B , Murray T, Thun M . Cancer statistics, 2001. C A Cancer J C l i n 2001; 51(l):15-36. 66. Guo Y , Fang Y , Huang B . [Study of loss of heterozygosity at chromosome 1 l q l 3 in nasopharyngeal carcinoma]. Zhonghua Zhong L i u Z a Zh i 2001; 23(2): 132-4. 67. Haidar M A , Kantarjian H , Manshouri T, et al. A T M gene deletion in patients 115 with adult acute lymphoblastic leukemia. Cancer 2000; 88(5): 1057-62. 68. Hampton G M , Mannermaa A , Winquist R, et al. Loss of heterozygosity in sporadic human breast carcinoma: a common region between 1 lq22 and l l q23 .3 . Cancer Res 1994; 54(17):4586-9. 69. Han L , Wong D , Dhaka A , et al. Protein binding and signaling properties of R T N l suggest a unique effector function. Proc Nat l Acad Sci U S A 1997; 94(10):4954-9. 70. Hawley R S , Friend S H . Strange bedfellows in even stranger places: the role of A T M in meiotic cells, lymphocytes, tumors, and its functional links to p53. Genes Dev 1996; 10(19):2383-8. 71. Herbst R A , Gutzmer R, Matiaske F, et al. Identification of two distinct deletion targets at 1 lq23 in cutaneous malignant melanoma. Int J Cancer 1999; 80(2):205-9. 72. Hinds P W , Dowdy SF, Eaton E N , et al. Function of a human cyclin gene as an oncogene. Proc Natl Acad Sci U S A 1994; 91(2):709-13. 73. Hirano A , E m i M , Tsuneizumi M , et al. Al le l i c losses of loci at 3p25.1, 8p22, 13ql2, 17pl3.3, and 22ql3 correlate with postoperative recurrence in breast 116 cancer. C l i n Cancer Res 2001; 7(4):876-82. 74. Hoechtlen-Vollmar W , Menzel G , Bartl R, et al. Amplification of cyclin D l gene in multiple myeloma: clinical and prognostic relevance. B r J Haematol 2000; 109(l):30-8. 75. Hou L , Shi D , Tu S M , et al. Oral cancer progression and c-erbB-2\/neu proto-oncogene expression. Cancer Lett 1992; 65(3):215-20. 76. H u i A B , Lo K W , Leung SF, et al. Loss of heterozygosity on the long arm of chromosome 11 in nasopharyngeal carcinoma. Cancer Res 1996; 56(14):3225-9. 77. Iwasaki H . A possible tumor suppressor gene for parathyroid adenomas. Int Surg 1996; 81(l):71-6. 78. Izzo JG , Papadimitrakopoulou V A , L i X Q , et al. Dysregulated cyclin D l expression early in head and neck tumorigenesis: in vivo evidence for an association with subsequent gene amplification. Oncogene 1998; 17(18):2313-22. 79. Izzo JG , Papadimitrakopoulou V A , L i X Q , et al. Dysregulated cyclin D l expression early in head and neck tumorigenesis: in vivo evidence for an 117 association with subsequent gene amplification. Oncogene 1998; 17(18):2313-22. 80. Janson M , Larsson C, Werelius B , et al. Detailed physical map of human chromosomal region 1 lq l2 -13 shows high meiotic recombination rate around the M E N 1 locus. Proc Natl Acad Sci U S A 1991; 88(23):10609-13. 81. Jares P, Nadal A , Fernandez P L , et al. Disregulation of p l 6 M T S l \/ C D K 4 I protein and m R N A expression is associated with gene alterations in squamous-cell carcinoma of the larynx. Int J Cancer 1999; 81(5):705-11. 82. Jiang W , Kahn S M , Zhou P, et al. Overexpression o f cyclin D l in rat fibroblasts causes abnormalities in growth control, cell cycle progression and gene expression. Oncogene 1993; 8(12):3447-57. 83. J in Y , Hoglund M , Jin C, et al. F I S H characterization o f head and neck carcinomas reveals that amplification of band 1 l q l 3 is associated with deletion of distal l l q . Genes Chromosomes Cancer 1998; 22(4):312-20. 84. Kaltz-Wittmer C, Klenk U , Glaessgen A , et al. F I S H analysis of gene aberrations ( M Y C , C C N D 1 , E R B B 2 , R B , and A R ) in advanced prostatic carcinomas before and after androgen deprivation therapy. Lab Invest 2000; 80(9): 1455-64. 118 85. Kanno H , Kondo K , Ito S, et al. Somatic mutations of the von Hippel-Lindau tumor suppressor gene in sporadic central nervous system hemangioblastomas. Cancer Res 1994; 54(18):4845-7. 86. Kastan M B , Zhan Q, el-Deiry W S , et al. A mammalian cell cycle checkpoint pathway utilizing p53 and G A D D 4 5 is defective in ataxia-telangiectasia. Ce l l 1992; 71(4):587-97. 87. Keegan K S , Holtzman D A , Plug A W , et al. The Atr and A t m protein kinases associate with different sites along meiotically pairing chromosomes. Genes Dev 1996; 10(19):2423-37. 88. Kiaris H , Spandidos D A , Jones A S , et al. Mutations, expression and genomic instability of the H-ras proto-oncogene in squamous cell carcinomas of the head and neck. B r J Cancer 1995; 72(1): 123-8. 89. K i m S K , Fan Y , Papadimitrakopoulou V , et al. D P C 4 , a candidate tumor suppressor gene, is altered infrequently in head and neck squamous cell carcinoma. Cancer Res 1996; 56(11):2519-21. 90. Kitagawa Y , Ueda M , Ando N , et al. Significance of int-2\/hst-l coamplification as a prognostic factor in patients with esophageal squamous carcinoma. Cancer Res 1991; 51(5):1504-8. 119 91. Knudson A G . Hereditary cancer, oncogenes, and antioncogenes. Cancer Res 1985; 45(4): 1437-43. 92. Knudson A G , Jr. Genetics of human cancer. J Ce l l Physiol Suppl 1986; 4:7-11. 93. Koike M , Takeuchi S, Park S, et al. Ovarian cancer: loss of heterozygosity frequently occurs in the A T M gene, but structural alterations do not occur in this gene. Oncology 1999; 56(2):160-3. 94. Koreth J, Bakkenist CJ , McGee JO. Al le l i c deletions at chromosome 1 lq22-q23.1 and 1 lq25-qterm are frequent in sporadic breast but not colorectal cancers. Oncogene 1997; 14(4):431-7. 95. Largey JS, Meltzer SJ, Sauk JJ, et al. Loss of heterozygosity involving the A P C gene in oral squamous cell carcinomas. Oral Surg Oral M e d Oral Pathol 1994; 77(3):260-3. 96. Larsson C, Skogseid B , Oberg K , et al. Multiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature 1988; 332(6159):85-7. 97. Lat i f F, Tory K , Gnarra J, et al. Identification of the von Hippel-Lindau 120 disease tumor suppressor gene. Science 1993; 260(5112):1317-20. 98. Lat i l A , Baron JC, Cussenot O, et al. Oncogene amplifications in early-stage human prostate carcinomas. Int J Cancer 1994; 59(5):637-8. 99. Launonen V , Stenback F , Puistola U , et al. Chromosome 1 Iq22.3-q25 L O H in ovarian cancer: association with a more aggressive disease course and involved subregions. Gynecol Oncol 1998; 71(2):299-304. 100. Launonen V , Laake K , Huusko P, et al. European multicenter study on L O H o f A P O C 3 at 1 lq23 in 766 breast cancer patients: relation to clinical variables. Breast Cancer Somatic Genetics Consortium. B r J Cancer 1999; 80(5-6):879-82. 101. Lazar A D , Winter M R , Nogueira C P , et al. Loss of heterozygosity at 1 lq23 in squamous cell carcinoma of the head and neck is associated with recurrent disease. C l i n Cancer Res 1998; 4(ll) :2787-93. 102. Leblanc T, Le Coniat M , Flexor M , et al. A n interstitial 1 lq23 deletion proven to be a rearrangement interrupting the M L L gene in an infant with acute myeloblastic leukemia. Leukemia 1996; 10(11): 1844-6. 103. Lengauer C, Kinzler K W , Vogelstein B . Genetic instabilities in human 121 cancers. Nature 1998; 396(6712):643-9. 104. Lens D , Matutes E , Catovsky D , Coignet L J . Frequent deletions at 1 lq23 and 13ql4 in B cell prolymphocyte leukemia ( B - P L L ) . Leukemia 2000; 14(3):427-30. 105. Lese C M , Rossie K M , Appel B N , et al. Visualization o f INT2 and HST1 amplification in oral squamous cell carcinomas. Genes Chromosomes Cancer 1995; 12(4):288-95. 106. L i X , Lee N K , Y e Y W , et al. Al le l i c loss at chromosomes 3p, 8p, 13q, and 17p associated with poor prognosis in head and neck cancer. J Natl Cancer Inst 1994; 86(20): 1524-9. 107. Liggett W H , Sewell D A , Rocco J, et al. p l 6 and p l 6 beta are potent growth suppressors of head and neck squamous carcinoma cells in vitro. Cancer Res 1996; 56(18):4119-23. 108. Lippman S M , Hong W K . Second malignant tumors in head and neck squamous cell carcinoma: the overshadowing threat for patients with early-stage disease. Int J Radiat Oncol B i o l Phys 1989; 17(3):691-4. 109. Lippman S M , Hong W K . Molecular markers of the risk o f oral cancer. N Engl 122 J M e d 2001;344(17):1323-6. 110. Lubensky IA, Debelenko L V , Zhuang Z , et al. A l l e l i c deletions on chromosome 1 l q l 3 in multiple tumors from individual M E N 1 patients. Cancer Res 1996; 56(22):5272-8. 111. Lumerman H , Freedman P, Kerpel S. Oral epithelial dysplasia and the development o f invasive squamous cell carcinoma. Oral Surg Oral M e d Oral Pathol Oral Radiol Endod 1995; 79(3):321-9. 112. Mao L , Fan Y H , Lotan R, Hong W K . Frequent abnormalities of FHIT , a candidate tumor suppressor gene, in head and neck cancer cell lines. Cancer Res 1996; 56(22):5128-31. 113. Mao E J , Schwartz S M , Daling JR, Beckmann A M . Loss o f heterozygosity at 5q21-22 (adenomatous polyposis coli gene region) in oral squamous cell carcinoma is common and correlated with advanced disease. J Oral Pathol M e d 1998; 27(7):297-302. 114. Mao L . Tumor suppressor genes: does FHIT fit? J Nat l Cancer Inst 1998; 90(6):412-4. 115. Mashberg A , Meyers H . Anatomical site and size of 222 early asymptomatic 123 oral squamous cell carcinomas: a continuing prospective study of oral cancer. II. Cancer 1976; 37(5):2149-57. 116. Mashberg A . Erythroplasia vs. leukoplasia in the diagnosis of early asymptomatic oral squamous carcinoma. N Engl J M e d 1977; 297(2): 109-10. 117. Masuda M , Hirakawa N , Nakashima T, et al. Cyc l in D l overexpression in primary hypopharyngeal carcinomas. Cancer 1996; 78(3):390-5. 118. Matsuda H , Konishi N , Hiasa Y , et al. Alterations of p l 6 \/ C D K N 2 , p53 and ras genes in oral squamous cell carcinomas and premalignant lesions. J Oral Pathol M e d 1996; 25(5):232-8. 119. McGrory JE, Pritchard D J , Arndt C A , et al. Nonrhabdomyosarcoma soft tissue sarcomas in children. The Mayo Cl inic experience. C l i n Orthop 2000(374):247-58. 120. M c K a y J A , Douglas JJ, Ross V G , et al. Cyc l in D l protein expression and gene polymorphism in colorectal cancer. Aberdeen Colorectal Initiative. Int J Cancer 2000; 88(1):77-81. 121. Meredith SD, Levine P A , Burns J A , et al. Chromosome 1 l q l 3 amplification in head and neck squamous cell carcinoma. Association with poor prognosis. 124 Arch Otolaryngol Head Neck Surg 1995; 121(7):790-4. 122. M e y n M S , Strasfeld L , Al l en C. Testing the role of p53 in the expression of genetic instability and apoptosis in ataxia-telangiectasia. Int J Radiat B i o l 1994; 66(6 Suppl):S141-9. 123. M e y n M S . Ataxia-telangiectasia and cellular responses to D N A damage. Cancer Res 1995; 55(24):5991-6001. 124. Michaux L , Wlodarska I, Stul M , et al. M L L amplification in myeloid leukemias: A study of 14 cases with multiple copies o f 1 lq23. Genes Chromosomes Cancer 2000; 29(l):40-7. 125. Morrel l D , Cromartie E , Swift M . Mortality and cancer incidence in 263 patients with ataxia-telangiectasia. J Natl Cancer Inst 1986; 77(l):89-92. 126. Mugica-Van Herckenrode C, Rodriguez J A , Iriarte-Campo V , et al. Definition of a region of loss of heterozygosity at chromosome 1 l q in cervical carcinoma. Diagn M o l Pathol 1999; 8(2):92-6. 127. Mul ler W J , Lee FS , Dickson C, et al. The int-2 gene product acts as an epithelial growth factor in transgenic mice. Embo J 1990; 9(3):907-13. 125 128. Mul ler D , M i l l o n R, Velten M , et al. Amplification of 1 l q l 3 D N A markers in head and neck squamous cell carcinomas: correlation with clinical outcome. Eur J Cancer 1997; 33(13):2203-10. 129. Murakami Y , Nobukuni T, Tamura K , et al. Localization o f tumor suppressor activity important in nonsmall cell lung carcinoma on chromosome l l q . Proc Natl Acad Sci U S A 1998; 95(14):8153-8. 130. Nakamura Y , Larsson C, Julier C, et al. Localization of the genetic defect in multiple endocrine neoplasia type 1 within a small region o f chromosome 11. A m J Hum Genet 1989; 44(5):751-5. 131. Nawroz H , van der Riet P, Hruban R H , et al. Allelotype o f head and neck squamous cell carcinoma. Cancer Res 1994; 54(5):1152-5. 132. Negrini M , Castagnoli A , Sabbioni S, et al. Suppression of tumorigenesis by the breast cancer cell line M C F - 7 following transfer of a normal human chromosome 11. Oncogene 1992; 7(10):2013-8. 133. Nord B , Larsson C, Wong F K , et al. Sporadic follicular thyroid tumors show loss of a 200-kb region in 1 l q l 3 without evidence for mutations in the M E N 1 gene. Genes Chromosomes Cancer 1999; 26(l):35-9. 126 134. Nowel l P C . The clonal evolution of tumor cell populations. Science 1976; 194(4260):23-8. 135. Nunn J, Scholes A G , Li loglou T, et al. Fractional allele loss indicates distinct genetic populations in the development of squamous cell carcinoma of the head and neck ( S C C H N ) . Carcinogenesis 1999; 20(12):2219-28. 136. Ornitz D M , Moreadith R W , Leder P. Binary system for regulating transgene expression in mice: targeting int-2 gene expression with yeast G A L 4 \/ U A S control elements. Proc Natl Acad Sci U S A 1991; 88(3):698-702. 137. Ott C E , Skroch E , Steinhart H , et al. Thin section arrays for I -FISH analysis of chromosome-specific imbalances in squamous cell carcinomas of the head and neck. Int J Oncol 2002; 20(3):623-30. 138. Papadimitrakopoulou V , Izzo J, Lippman S M , et al. Frequent inactivation of p l 6 I N K 4 a in oral premalignant lesions. Oncogene 1997; 14(15):1799-803. 139. Papadimitrakopoulou V A , Hong W K . Retinoids in head and neck chemoprevention. Proc Soc Exp B i o l M e d 1997; 216(2):283-90. 140. Partridge M , Emil ion G , Pateromichelakis S, et al. A l l e l i c imbalance at chromosomal loci implicated in the pathogenesis o f oral precancer, 127 cumulative loss and its relationship with progression to cancer. Oral Oncol 1998; 34(2):77-83. 141. Partridge M , Emil ion G , Pateromichelakis S, et al. Location of candidate tumour suppressor gene loci at chromosomes 3p, 8p and 9p for oral squamous cell carcinomas, hit J Cancer 1999; 83(3):318-25. 142. Partridge M , Pateromichelakis S, Phillips E , et al. Profiling clonality and progression in multiple premalignant and malignant oral lesions identifies a subgroup of cases with a distinct presentation of squamous cell carcinoma. C l i n Cancer Res 2001; 7(7): 1860-6. 143. Paterson IC, Eveson JW, Prime SS. Molecular changes in oral cancer may reflect aetiology and ethnic origin. Eur J Cancer B Oral Oncol 1996; 32B(3): 150-3. 144. Pavelic Z P , Gluckman JL . The role of p53 tumor suppressor gene in human head and neck tumorigenesis. Acta Otolaryngol Suppl 1997; 527:21-4. 145. Pinto A P , Crum C P . Natural history of cervical neoplasia: defining progression and its consequence. C l i n Obstet Gynecol 2000; 43(2):352-62. 146. Pruneri G , Fabris S, Baldini L , et al. Immunohistochemical analysis of cyclin 128 D l shows deregulated expression in multiple myeloma with the t ( l 1; 14). A m J Pathol 2000; 156(5):1505-13. 147. Pulido H A , Fakruddin M J , Charter] ee A , et al. Identification of a 6 - cM minimal deletion at l lq23.1-23.2 and exclusion of P P P 2 R 1 B gene as a deletion target in cervical cancer. Cancer Res 2000; 60(23):6677-82. 148. Quelle D E , Ashmun R A , Shurtleff S A , et al. Overexpression of mouse D-type cyclins accelerates G l phase in rodent fibroblasts. Genes Dev 1993; 7(8):1559-71. 149. Rao B R , Krishnamurthy P. A comparative study of the cost and effectiveness of a modified system of M D T drug delivery system in a high endemic district (Nalagonda) of South India. Indian J Lepr 1998; 70(Suppl(l)):63S-71S. 150. Reddy K S , Parsons L , M a k L , et al. Segmental amplification of 1 lq23 region identified by fluorescence in situ hybridization in four patients with myeloid disorders: a review. Cancer Genet Cytogenet 2001; 126(2): 139-46. 151. Reed A L , Califano J, Cairns P, et al. High frequency of p 16 ( C D K N 2 \/ M T S -1\/INK4A) inactivation in head and neck squamous cell carcinoma. Cancer Res 1996; 56(16):3630-3. 129 152. Reissmann PT, Koga H , F ig l in R A , et al. Amplification and overexpression of the cyclin D l and epidermal growth factor receptor genes in non-small-cell lung cancer. Lung Cancer Study Group. J Cancer Res C l i n Oncol 1999; 125(2):61-70. 153. Renan M J . How many mutations are required for tumorigenesis? Implications from human cancer data. M o l Carcinog 1993; 7(3): 139-46. 154. Rennstam K , Baldetorp B , Kytola S, et al. Chromosomal rearrangements and oncogene amplification precede aneuploidization in the genetic evolution of breast cancer. Cancer Res 2001; 61(3):1214-9. 155. Riviere A , Wilckens C , Loning T. Expression of c-erbB2 and c-myc in squamous epithelia and squamous cell carcinomas of the head and neck and the lower female genital tract. J Oral Pathol M e d 1990; 19(9):408-13. 156. Robertson G , Coleman A , Lugo T G . A malignant melanoma tumor suppressor on human chromosome 11. Cancer Res 1996; 56(19):4487-92. 157. Rodrigo JP, Garcia L A , Ramos S, et al. E M S 1 gene amplification correlates with poor prognosis in squamous cell carcinomas o f the head and neck. C l i n Cancer Res 2000; 6(8):3177-82. 130 158. Roelofs H , Schuuring E , Wiegant J, et al. Amplification o f the 1 l q l 3 region in human carcinoma cell lines: a mechanistic view. Genes Chromosomes Cancer 1993; 7(2):74-84. 159. Roh H J , Shin D M , Lee JS, et al. Visualization of the timing of gene amplification during multistep head and neck tumorigenesis. Cancer Res 2000; 60(22):6496-502. 160. Rosin M P , Epstein JB , Berean K , et al. The use o f exfoliative cell samples to map clonal genetic alterations in the oral epithelium of high-risk patients. Cancer Res 1997; 57(23):5258-60. 161. Rosin M P , Cheng X , Poh C, et al. Use of allelic loss to predict malignant risk for low-grade oral epithelial dysplasia. C l i n Cancer Res 2000; 6(2):357-62. 162. Rubin JS, Qiu L , Etkind P. Amplification of the Int-2 gene in head and neck squamous cell carcinoma. J Laryngol Otol 1995; 109(l):72-6. 163. Sakata K . Alterations of tumor suppressor genes and the H-ras oncogene in oral squamous cell carcinoma. J Oral Pathol M e d 1996; 25(6):302-7. 164. Saranath D , Bhoite L T , Deo M G . Molecular lesions in human oral cancer: the Indian scene. Eur J Cancer B Oral Oncol 1993; 29B(2): 107-12. 131 165. Sartor M , Steingrimsdottir H , Elamin F , et al. Role o f p l 6 \/ M T S l , cyclin D l and R B in primary oral cancer and oral cancer cell lines. B r J Cancer 1999; 80(l-2):79-86. 166. Schell H , Schonberger A . [Site-specific incidence o f benign and precancerous leukoplakias and cancers of the oral cavity]. Z Hautkr 1987; 62(10):798-804. 167. Schepman K P , van der Meij E H , Smeele L E , van der Waal I. Malignant transformation of oral leukoplakia: a follow-up study o f a hospital-based population of 166 patients with oral leukoplakia from The Netherlands. Oral Oncol 1998; 34(4):270-5. 168. Schuuring E , Verhoeven E , M o o i W J , Michalides R J . Identification and cloning of two overexpressed genes, U21B31 \/PRAD1 and E M S 1 , within the amplified chromosome l l q l 3 region inhuman carcinomas. Oncogene 1992; 7(2):355-61. 169. Schuuring E , Verhoeven E , Litvinov S, Michalides R J . The product of the E M S 1 gene, amplified and overexpressed in human carcinomas, is homologous to a v-src substrate and is located in cell-substratum contact sites. M o l Ce l l B i o l 1993; 13(5):2891-98. 170. Schuuring E . The involvement of the chromosome 1 l q l 3 region in human 132 malignancies: cyclin D l and E M S 1 are two new candidate oncogenes~a review. Gene 1995; 159(l):83-96. 171. Schweigerer L . Fibroblast growth factor and angiogenesis. Z Kardiol 1989; 78(Suppl 6): 12-5. 172. Shaffnan T, Khanna K K , Kedar P, et al. Interaction between A T M protein and c -Ab l in response to D N A damage. Nature 1997; 387(6632):520-3. 173. Shekhar M P , Nangia-Makker P, Wolman SR, et al. Direct action of estrogen on sequence of progression of human preneoplastic breast disease. A m J Pathol 1998; 152(5):1129-32. 174. Shikhani A H , Matanoski G M , Jones M M , et al. Mult iple primary malignancies in head and neck cancer. A r c h Otolaryngol Head Neck Surg 1986; 112(ll):1172-9. 175. Shuster M I , Han L , Le Beau M M , et al. A consistent pattern o f RIN1 rearrangements in oral squamous cell carcinoma cell lines supports a breakage-fusion-bridge cycle model for l l q l 3 amplification. Genes Chromosomes Cancer 2000; 28(2):153-63. 176. Sidransky D . Nucleic acid-based methods for the detection o f cancer. Science 133 1997;278(5340):1054-9. 177. Silverman S, Gorsky M , Lozada F. Oral leukoplakia and malignant transformation. A follow-up study o f 257 patients. Cancer 1984; 53(3):563-8. 178. Skomedal H , Helland A , Kristensen G B , et al. Al le l i c imbalance at chromosome region 1 lq23 in cervical carcinomas. Eur J Cancer 1999; 35(4):659-63. 179. Sparks A B , M o r i n PJ , Vogelstein B , Kinzler K W . Mutational analysis of the APC\/beta-catenin\/Tcf pathway in colorectal cancer. Cancer Res 1998; 58(6):1130-4. 180. Steenbergen R D , Hermsen M A , Walboomers J M , et al. Integrated human papillomavirus type 16 and loss o f heterozygosity at l l q 2 2 and 18q21 in an oral carcinoma and its derivative cell line. Cancer Res 1995; 55(22):5465-71. 181. Streubel B , Valent P, Jager U , et al. Amplification o f the M L L gene on double minutes, a homogeneously staining region, and ring chromosomes in five patients with acute myeloid leukemia or myelodysplastic syndrome. Genes Chromosomes Cancer 2000; 27(4):380-6. 182. Sugimura T, Yoshida T, Sakamoto H , et al. Molecular biology of the hst-1 134 gene. Ciba Found Symp 1990; 150:79-89. 183. Tada K , Oka M , Tangoku A , et al. Gains of 8q23-qter and 20q and loss of 1 lq22-qter in esophageal squamous cell carcinoma associated with lymph node metastasis. Cancer 2000; 88(2):268-73. 184. Taylor A M . Ataxia telangiectasia genes and predisposition to leukaemia, lymphoma and breast cancer. B r J Cancer 1992; 66(l):5-9. 185. Telatar M , Wang S, Castellvi-Bel S, et al. A model for A T M heterozygote identification in a large population: four founder-effect A T M mutations identify most of Costa Rican patients with ataxia telangiectasia. M o l Genet Metab 1998; 64(l):36-43. 186. Terada M , Sakamoto H , Yoshida T, et al. A novel transforming gene, hst, from human stomach cancers and a non-cancerous portion of stomach mucosa. Princess Takamatsu Symp 1986; 17:123-31. 187. Thakker R V , Bouloux P, Wooding C, et al. Association o f parathyroid tumors in multiple endocrine neoplasia type 1 with loss of alleles on chromosome 11. N Engl J M e d 1989; 321(4):218-24. 188. Todd R, McBride J, Tsuji T, et al. Deleted in oral cancer-1 (doc-1), a novel 135 oral tumor suppressor gene. Faseb J 1995; 9(13):1362-70. 189. Tomlinson I, Ilyas M , Novel l i M . Molecular genetics of colon cancer. Cancer Metastasis Rev 1997; 16(l-2):67-79. 190. Uchida A , Tachibana M , Miyakawa A , et al. Microsatellite analysis in multiple chromosomal regions as a prognostic indicator o f primary bladder cancer. Uro l Res 2000; 28(5):297-303. 191. Uchimaru K , Endo K , Fujinuma H , et al. Oncogenic collaboration of the cyclin D l ( P R A D 1 , bcl-1) gene with a mutated p53 and an activated ras oncogene in neoplastic transformation. Jpn J Cancer Res 1996; 87(5):459-65. 192. Uzawa K , Yoshida H , Suzuki H , et al. Abnormalities of the adenomatous polyposis coli gene in human oral squamous-cell carcinoma. Int J Cancer 1994; 58(6):814-7. 193. Uzawa K , Suzuki H , Komiya A , et al. Evidence for two distinct tumor-suppressor gene loci on the long arm of chromosome 11 in human oral cancer. Int J Cancer 1996; 67(4):510-4. 194. van der Waal I, Schepman K P , van der Meij E H , Smeele L E . Oral leukoplakia: a clinicopathological review. Oral Oncol 1997; 33(5):291-301. 136 195. Venugopalan M , Wood TF , Wilczynski SP, et al. Loss o f heterozygosity in squamous cell carcinomas of the head and neck defines a tumor suppressor gene region on l l q l 3 . Cancer Genet Cytogenet 1998; 104(2): 124-32. 196. Vogelstein B , Fearon E R , Hamilton SR, et al. Genetic alterations during colorectal-tumor development. N Engl J M e d 1988; 319(9):525-32. 197. Vogelstein B , Kinzler K W . The genetic basis o f human cancer, 1998. 198. Waber P G , Lee N K , Nisen P D . Frequent allelic loss at chromosome arm 3p is distinct from genetic alterations of the Von-Hippel Lindau tumor suppressor gene in head and neck cancer. Oncogene 1996; 12(2):365-9. 199. Waha A , Sturne C, Kessler A , et al. Expression of the A T M gene is significantly reduced in sporadic breast carcinomas. Int J Cancer 1998; 78(3):306-9. 200. Waldron C A , Shafer W G . Leukoplakia revisited. A clinicopathologic study 3256 oral leukoplakias. Cancer 1975; 36(4): 1386-92. 201. Wang T C , Cardiff R D , Zukerberg L , et al. Mammary hyperplasia and carcinoma in M M T V - c y c l i n D l transgenic mice. Nature 1994; 369(6482):669-71. 137 202. Wang M B , A l a v i S, Engstrom M , et al. Detection of chromosome 1 l q l 3 amplification in head and neck cancer using fluorescence in situ hybridization. Anticancer Res 1999; 19(2A):925-31. 203. Wang SS, Virmani A , Gazdar A F , et al. Refined mapping o f two regions of loss of heterozygosity on chromosome band 1 lq23 in lung cancer. Genes Chromosomes Cancer 1999; 25(2): 154-9. 204. Warnakulasuriya K A , Chang SE, Johnson N W . Point mutations in the Ha-ras oncogene are detectable in formalin-fixed tissues of oral squamous cell carcinomas, but are infrequent in British cases. J Oral Pathol M e d 1992; 21(5):225-9. 205. Weber JL, M a y P E . Abundant class of human D N A polymorphisms which can be typed using the polymerase chain reaction. A m J H u m Genet 1989; 44(3):388-96. 206. Weber G , Friedman E , Grimmond S, et al. The phospholipase C beta 3 gene located in the M E N 1 region shows loss of expression in endocrine tumours. Hum M o l Genet 1994; 3(10): 1775-81. 207. Weitzel J N , Patel J, Smith D M , et al. Molecular genetic changes associated with ovarian cancer. Gynecol Oncol 1994; 55(2):245-52. 138 208. Westphal C H , Rowan S, Schmaltz C, et al. atm and p53 cooperate in apoptosis and suppression of tumorigenesis, but not in resistance to acute radiation toxicity. Nat Genet 1997; 16(4):397-401. 209. Will iams M E , Gaffey M J , Weiss L M , et al. Chromosome 11Q13 amplification in head and neck squamous cell carcinoma. A r c h Otolaryngol Head Neck Surg 1993; 119(11): 1238-43. 210. Wong A J , Croce C M . Oncogenes and signal transduction. Hosp Pract (Off Ed) 1993; 28(7):128-32, 136-8, 141. 211. X u L , Meng Y , Wallen R, DePinho R A . Loss o f transcriptional attenuation in N-myc is associated with progression towards a more malignant phenotype. Oncogene 1995; 11(9): 1865-72. 212. X u G , Livingston D M , Krek W . Multiple members of the E2F transcription factor family are the products of oncogenes. Proc Nat l Acad Sci U S A 1995; 92(5):1357-61. 213. X u J, Gimenez-Conti IB, Cunningham JE, et al. Alterations of p53, cyclin D l , Rb, and H-ras in human oral carcinomas related to tobacco use. Cancer 1998; 83(2):204-12. 139 214. Yokoyama J, Shiga K , Sasano H , et al. Abnormalities and the implication of retinoblastoma locus and its protein product in head and neck cancers. Anticancer Res 1996; 16(2):641-4. 215. Yoshida M C , Wada M , Satoh H , et al. Human HST1 (HSTF1) gene maps to chromosome band 1 l q l 3 and coamplifies with the INT2 gene in human cancer. Proc Natl Acad Sci U S A 1988; 85(13):4861-4. 216. Zhang L , Michelsen C, Cheng X , et al. Molecular analysis o f oral lichen planus. A premalignant lesion? A m J Pathol 1997; 151(2):323-7. 217. Zhou P, Jiang W , Weghorst C M , Weinstein IB. Overexpression of cyclin D l enhances gene amplification. Cancer Res 1996; 56(l):36-9. 218. Zhu Y , Monni O, El -Rifa i W , et al. Discontinuous deletions at 1 lq23 in B cell chronic lymphocytic leukemia. Leukemia 1999; 13(5):708-12. 219. Zhu Y , Monni O, Franssila K , et al. Deletions at 1 lq23 in different lymphoma subtypes. Haematologica 2000; 85(9):908-12. 220. Zhuang Z , Merino M J , Chuaqui R, et al. Identical allelic loss on chromosome 1 l q l 3 in microdissected in situ and invasive human breast cancer. Cancer Res 1995;55(3):467-71. 140 ","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/hasType":[{"value":"Thesis\/Dissertation","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#dateIssued":[{"value":"2002-11","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt":[{"value":"10.14288\/1.0090340","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/language":[{"value":"eng","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline":[{"value":"Dental Science","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/provider":[{"value":"Vancouver : University of British Columbia Library","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/publisher":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/rights":[{"value":"For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https:\/\/open.library.ubc.ca\/terms_of_use.","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#scholarLevel":[{"value":"Graduate","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/title":[{"value":"Allelic imbalance at 11q in oral cancer and premalignant lesions","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/type":[{"value":"Text","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#identifierURI":[{"value":"http:\/\/hdl.handle.net\/2429\/12316","type":"literal","lang":"en"}]}}