Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Malignant and Nonmalignant Gene Signatures in Squamous Head and Neck Cancer

Malignant and Nonmalignant Gene Signatures in Squamous Head and Neck Cancer Hindawi Publishing Corporation Journal of Oncology Volume 2012, Article ID 752860, 8 pages doi:10.1155/2012/752860 Research Article Malignant and Nonmalignant Gene Signatures in Squamous Head and Neck Cancer 1 2 1 1 Maria J. Worsham, Mei Lu, Kang Mei Chen, Josena K. Stephen, 1 1 Shaleta Havard, and Vanessa P. Schweitzer Department of Otolaryngology-Head & Neck Surgery, Henry Ford Health System, Detroit, MI 48202, USA Department of Biostatistics and Research Epidemiology, Henry Ford Health System, Detroit, MI 48202, USA Correspondence should be addressed to Maria J. Worsham, mworsha1@hfhs.org Received 7 December 2011; Revised 26 January 2012; Accepted 13 February 2012 Academic Editor: Mario A. Hermsen Copyright © 2012 Maria J. Worsham et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Genetic events specific to the pathogenesis of malignancy can offer clues to the tumorigenesis process. The objective of this study was to identify gene alterations that differentiate tumor and nontumor lesions in squamous head and neck cancer (HNSCC). DNA from 220 primary HNSCC with concurrently present tumor and nontumor lesions from the same patient was interrogated for ge- nomic alterations of loss or gain of copy. Conditional logistic regression dealt with tumor and non-tumor records within a patient. Of 113 genes, 53 had univariate effects (P< 0.01), of which 16 genes remained in the multivariable model withP< 0.01. The model had a C-index (ROC) of 0.93. Loss of CDKN2B and gain of BCL6, FGF3,and PTP4A3 predicted tumor. Loss of BAK1 and CCND1 and gain of STCH predicted nontumor. This highly powered model assigned alterations in 16 genes specific for malignant versus nonmalignant lesions, supporting their contribution to the pathogenesis of HNSCC as well as their potential utility as relevant tar- gets for further evaluation as markers of early detection and progression. 1. Introduction cancer genes in many tumor types; however, individual gene loci altered in tumors cannot be deduced solely from the type Knowledge of the genetic mechanisms that drive cancer of chromosome rearrangement [1]. Historically, the molecu- growth and development is important in understanding the lar pathogenesis of cancer has been teased out one gene at a pathogenesis of malignancy and provides insights into the time. Recent high-throughput genomewide candidate strat- tumorigenesis process. The underlying hypothesis is that be- egies such as the multiplex ligation-dependent probe ampli- havior of tumor cells is determined by genetic changes that fication (MLPA) assay [2] to identify specific genes for gain alter cell growth, cell differentiation, programmed cell death, and loss concurred with chromosomal aberrations and pro- and cell migration. Cancer is the result of transformation vide a novel index to estimate the extent of genomic abnor- from a normal to a malignant cell that results from accumu- mality with disease progression [1]. lated mutations. Acquisition of a fully malignant phenotype Molecular genetic prognosticators can influence preven- in colon cancer is thought to occur because of multiple steps tion, diagnosis, appropriateness of adjuvant chemotherapy, whose targets are alterations of growth-promoting onco- and, possibly, the chemotherapeutic regimen of cancer pa- genes and growth-inhibiting cancer suppressor genes. The tients. Dissecting out processes specific to the pathogenesis evolution in transformation from a normal squamous epi- of malignancy can distill key genetic biomarkers of HNSCC thelial cell to a cancer cell is likewise assumed to require sev- etiology, transformation, and progression. eral steps, some defined by genetic alteration. Genetic alterations provide means of identifying tumor In this study, in a primary HNSCC cohort of 220 patients, cells as well as defining changes that presumably determine with both tumor and nontumor lesions within a biopsy (tu- biological differences from their normal counterparts. Chro- mor and nontumor from the same patient), we examined mosome aberrations have served as landmarks to identify gene alterationsoflossand gain to derive multivariate 2 Journal of Oncology predictive models to discriminate malignant from nonma- reactions (probe sets p005, p006, p007, MRC Holland, Am- lignant lesions. sterdam). Briefly, DNA, diluted with water to a total volume of 5 μL, is denatured and fragmented by heating for 5 min- utes at 98 C in a thermocycler. Binary MLPA probes are add- 2. Methods ed and allowed to hybridize to their targets during an 16 hr incubation at 60 C, followed by the addition of dilution buf- 2.1. Patient Cohort. Cohort subjects were male and female fer and a ligase enzyme (MRC Holland, Amsterdam). During patients 18 years and older who underwent a HNSCC tissue the 15-minute incubation at 60 C, the two parts of a probe biopsy at the Henry Ford Health System from 1986–2006. become ligated to each other and become an amplifiable The use of formalin-fixed paraffin-embedded tissue blocks molecule if the complementary sequence is present in from patients with both tumor and nontumor records within the DNA sample. This is followed by the addition of PCR pri- the same biopsy and the collection of related patient infor- mers, dNTPs and Taq polymerase, followed by the following mation were approved by the Henry Ford Health System In- ◦ ◦ ◦ cycles: one at 1 min 95 C, 10 cycles: 30 sec 95 C, 30 sec 70 C, stitutional Review Board (IRB) Committee. ◦ ◦ ◦ ◦ 1min 72 C; 30 cycles at 30 sec 95 C, 30 sec 60 C, 1 min 72 C. In the event a study subject had more than one eligible Thesameprimerpair, oneofwhich is tagged with afluores- biopsy over the course of the study period, the primary bi- cent dye, amplifies all (ligated) probes. The relative amounts opsy was the index biopsy and the pathology report date marked the study enrollment (index) date. Cohort subjects of PCR product obtained reflect the relative amounts of ligat- missing biopsy tissue blocks or insufficient tissue for molec- ed probes at the start of the PCR reaction. Amplification pro- ular analysis were excluded. ducts are analyzed on a DNA sequencer (Applied Biosystems, Foster City, Ca), quantified and interpreted as previously des- cribed [1, 3, 4, 6–8]. 2.2. Histopathology. Pathology review of paraffin-embedded tissue sections captured all lesion types in a biopsy to include normal squamous epithelium, squamous dysplasia whether 2.4. Statistical Analysis. Conditional logistic regression mod- classified as mild, moderate, or severe/carcinoma in situ, and eling was used to address tumor and nontumor lesions with- tumor. Severe dysplasia and carcinoma in situ lesions were in the primary biopsy in an HNSCC patient. Analysis began groupedwithtumor andconsideredasmalignant outcomes. by testing individual genes as risk predictors/discriminators for tumor and nontumor (univariate analysis). Genes with 2.3. Molecular Analysis individual risks in a univariate analysis (P< 0.01) were can- didates for the first multivariable model. Prior to multivari- 2.3.1. Processing Lesion Specimens for Molecular Analysis. able modeling, genes were evaluated for their correlation and DNA was obtained from either whole 5 micron tissue sec- missing values. Highly correlated genes (correlation coeffi- tions (if the tissue block contained predominantly tumor or cient [r] > 0.7) or genes with larger missing values (>5%) nontumor tissue) or from microdissected tissue as previously were fitted separately along with other uncorrelated (r < 0.7) described [3]. Briefly, concurrently present tumor (severe genes. The stepwise model selection was considered. The dysplasia, carcinoma in situ, tumor) and nontumor (normal, final model included genes with P< 0.01 along with odds mild/moderate dysplasia) lesions in the same paraffin-em- ratios for loss or gain as risk predictors. The C-index/ROC bedded formalin fixed tissue block were marked by the path- (the receiver operating characteristic {ROC} curve), in a ologist and individual lesions were microdissected from 5 range of 0 to 1, is a measure of the model’s predictive ability, micron sections mounted on glass slides using a single-use where 0.5 indicates no discrimination and 0.7 or greater indi- disposable scalpel blade under a dissecting microscope. This procedure minimizes mixing of normal and tumor subpop- cates that the model is predictive. ulations and yields lesion and tumor samples estimated to be at least 90% free from contamination with normal cells 3. Results and Discussion [3, 4]. 3.1. Results. Matched tumor and nontumor lesions within 2.3.2. The Multiplex Ligation-Dependent Probe Amplification each patient in the 220 primary HNSCC cohort comprised (MLPA) Assay. MLPA has several advantages. It establishes a total of 1076 tissue records. There were 504 normal/mild/ the copy number of up to 41 nucleic acid sequences in one moderate dysplasia lesions (495: normal squamous epithe- single reaction. MLPA probes are able to discriminate bet- lium, 6: mild dysplasia, 3: moderate dysplasia), and 572 tu- ween sequences that differ in only one nucleotide. Moreover, mor lesions (568: tumor, 1: severe dysplasia, 3: carcinoma in MLPA reactions require a minimum of only 20 ng human situ (CIS)). Squamous mucosal dysplasia whether classified DNA making it especially amenable for PCR of DNA from as mild, moderate, or severe/carcinoma in situ is considered formalin-fixed paraffin-embedded head and neck tissues [3– intraepithelial neoplasia, and, as a precursor lesion group is 5]. separate from normal and tumor. The number of precursor, Starting with, approximately, 20–50 ng of genomic DNA, lesions is very small (13) in comparison to normal epithe- for each subject in the cohort, DNA from tumor and nontu- lium (495) and carcinomas (568). Of the 572 tumor lesions, mor is amplified for 122 probes (113 unique genes associated 193 (34%) were laryngeal, 170 (30%) were oral cavity, 151 with cancer including HNSCC) as separate sets of three (26%) were pharyngeal (oropharyngeal/hypopharyngeal), Journal of Oncology 3 Table 1: Genes alterations that predict malignant (M, bolded) and nonmalignant (NM, italicized). Effect Chromosome Odds ratio estimate Lower CL Upper CL §M CTNNB1 lossversusnormal 3p22 2.682 1.394 5.162 NM CTNNB1 gain versus normal 0.323 0.147 0.71 BCL6 loss versus normal 3q27 0.55 0.27 1.12 BCL6 gain versus normal 8.989 3.155 25.612 §M IL2 loss versus normal 4q26 3.697 1.774 7.705 NM IL2 gain versus normal 0.149 0.055 0.407 NM BAK1 lossversusnormal 6p21.3 0.262 0.103 0.666 BAK1 gain versus normal 0.438 0.192 0.999 §M LTA lossversusnormal 6p21.3 2.156 1.172 3.965 NM LTA gain versus normal 0.266 0.108 0.655 §NM FGFR1 lossversusnormal 8p21 0.275 0.126 0.598 FGFR1 gain versus normal 5.555 1.689 18.267 §NM PRKDC loss versus normal 8q11 0.276 0.11 0.692 PRKDC gain versus normal 5.449 2.09 14.206 §NM MYC loss versus normal 8q24.12 0.221 0.097 0.503 MYC gain versus normal 2.218 1.136 4.332 PTP4A3 loss versus normal 8q24.3 0.493 0.214 1.135 PTP4A3 gain versus normal 12.158 3.461 42.71 §M CDKN2A loss versus normal 9p21 1.845 1.013 3.362 NM CDKN2A gain versus normal 0.14 0.056 0.35 CDKN2B loss versus normal 9p21 3.256 1.676 6.325 CDKN2B gain versus normal 1.168 0.442 3.087 LMO2 loss versus normal 11p13 4.977 2.16 11.466 LMO2 gain versus normal 0.573 0.205 1.607 FGF3 loss versus normal 11q13 0.882 0.447 1.741 FGF3 gain versus normal 7.819 3.286 18.604 NM CCND1 loss versus normal 11q13 0.403 0.22 0.736 CCND1 gain versus normal 1.239 0.634 2.421 STCH loss versus normal 21q11.1 1.788 0.833 3.839 NM STCH gain versus normal 0.124 0.043 0.359 §M TFF1 loss versus normal 21q22.3 3.019 1.514 6.02 NM TFF1 gain versus normal 0.08 0.024 0.268 CL: confidence limit, : genes with loss and gain signifying aneuploidy. and 58 (10%) were lesions in the other category (nasophar- patients, 25% (55/220) were female and 75% (165/220) were ynx, nasal cavity, paranasal sinuses). male. The 1076 lesions were distributed among 932 tissue The missing value for each gene was in a range of 0% to blocks, of which 434 (47%) had tumor lesions only, 363 4.2%. Of the 113 unique genes (selected based on their asso- (39%) had nontumor lesions only, and 135 (14%) tissue ciation with cancer including HNSCC), 53 genes had uni- blocks had both tumor and nontumor lesions. Within a pa- variate effects (P< 0.01) and were considered as the candi- tient, the number of nontumor lesions ranged from 1 to 6 date genes for multivariable analyses. After the stepwise and from 1 to 7 for tumor lesions. Of the 220 patients with model selection, 16 genes remained in the multivariable both tumor and nontumor within the same patient, 35 model (P< 0.01) (Table 1). The model had a C-index (re- (16%) had 1 tumor and nontumor record, 28 (13%) had 2 ceiver operating characteristic (ROC)) of 0.93. tumor and nontumor records, 27 (12%) had 3 tumor and The 16 genes in the final model with alterations of loss nontumor records. and/or gain accounted for loci along 7 chromosomes: 3, 4, Fifty percent of cohort (110/220) subjects were Caucasian 6, 8, 9, 11, and 21 (Table 1). Of these, 50% were altered in American (CA), 38% (84/220) were self-reported as African both tumor and nontumor, with loss or gain reflective of American (AA), 4% (9/220) were non-CA/non-AA (His- chromosomal aneuploidy. This copy number instability panic: 2, Asian Pacific Islander: 2, Middle Eastern: 3, other: favored loss of CDKN2A (9p21), CTNNB1 (3p21), IL2 2), and, in 8% (17/220), race was missing. Of the 220 (4q26), LTA (6p21.3), and TFF1 (21q22.3) in tumor, with 4 Journal of Oncology corresponding gain in nontumor lesions, and gain of FGFR1 (8q24.12), and PRKDC (8q11) in tumor, with corresponding (8p21), MYC (8q24.12), and PRKDC (8q11) in tumor, with loss in nontumor. Chromosomal instability occurs early corresponding loss in nontumor (Table 1). along the tumorigenesis continuum and aneuploidy at the Loss of CDKN2B (9p21) and LMO2 (11p13) and gain of 9p21 locus affecting corresponding loss and gain in tumor BCL6 (3q27), FGF3 (11q13), and PTP4A3 (8q24.3) predicted and normal tissue, respectively, concurs with the proposed tumor. Loss of BAK1 (6p21.3) and CCND1 (11q13) and gain postulated model of molecular carcinogenesis for HNSCC of STCH (21q11.1) predicted nontumor (Table 1). [29]. In HNSCC, chromosomal aberrations on the long arm of Analysis excluding the 6 mild dysplasia and 3 moderate chromosome 3, resulting in gain of distal 3q segments, have dysplasia lesions from the nontumor group and the 1 severe been reported as recurring karyotypic alterations [22]. Gain dysplasia lesion and 3 CIS from the tumor group generated of 3q is supported by increased copy number (3-4 copies) of an identical multivariable model outcome. PIK3CA at 3q26.3, MME (3q25.1), and BCL6 genes at 3q27 [1] in HNSCC. In this study, gain of BCL6 was significantly 3.2. Discussion. Cancerous tissue in most cases has a dis- associated with tumor lesions. The protein encoded by BCL6 tinctive appearance under the microscope. Distinguishing is a zinc finger transcription factor and acts as a sequence- traits include a large number of dividing cells, variation in specific repressor of transcription. nuclear size and shape, variation in cell size and shape, loss Another chromosome 3 gene, CTNNB1 (catenin beta-1) of specialized cell features, loss of normal tissue organization, in the short arm at 3p21, is an adherens junction protein, and a poorly defined tumor boundary. Biopsy and micro- closely associated with adhesion, invasion, and metastasis in scopical examination can also distinguish malignancy, pre- different types of tumors, including SCC of the tongue [30]. cursor lesions of carcinoma in situ, mild, moderate, and The 3p21 region had the highest rate of allelic deletion (63%) severe dysplasia, and less reliably, hyperplasia, from normal in HNSCC [31]and is supportedbylossof CTNNB1 in tu- appearing tissue. mor lesions in this study. Corresponding gain of CTNNB1 In the multistep process of tumorigenesis, hyperplasia, copy number in nontumor lesions underscores chromoso- which refers to tissue growth based on an excessive rate of cell mal instability and ensuing aneuploidy as early events in the division, leading to a larger than usual number of cells but tumorigenesis process. with a normal orderly arrangement of cells within the tissue, Loss and corresponding gain of IL2 at 4q26 was signifi- and considered reversible, is thought to precede dysplasia. cantly associated with tumor and nontumor lesions, respec- Dysplasia, an abnormal type of excessive cell proliferation tively. The IL2 protein is produced by T cells in response to characterized by loss of normal tissue arrangement and cell antigenic or mitogenic stimulation and is required for T-cell structure, may revert to normal behavior, but, occasionally, proliferation and other activities crucial to regulation of the these lesions gradually become malignant. Distinguishing immune response. true precursor lesions on the basis of morphology alone is BAK1 (6p21.3) is a proapoptotic member of the BCL-2 often unreliable. In the molecular progression of HNSCC, family of genes that are involved in regulation of program- normal or minimal dysplasia often harbors abnormal geno- med cell death, and its increased expression had poorer dis- ease-specific survival in oral tongue squamous cell carcino- types [9], which do not necessarily correlate with observable changes in phenotype (morphology) [10, 11]. mas [32]. As a corollary to increased expression [32], in this In this study, 16 gene alterations with significant discrim- study, loss of BAK1 was a nontumor-specific event. Gene alterations were noted for 4 genes on chromosome inatory ability differentiated malignant HNSCC from non- malignant tissue. For matched tumor and nontumor lesions 8, three on 8q and one on 8p. Gains or amplifications involv- from the same patient within the 220 primary HNSCC study ing chromosome arm 8q are one of the most recurrent chro- cohort, examined for alterations in 113 unique genes with mosomal alterations in head and neck tumors. The human association to head and neck cancer, molecular alterations in protein tyrosine phosphatase type IVA, member 3, also known as PTP4A3, is located at 8q24.3 [33]. The protein en- “normal” appearing epithelium within the environment of a malignant biopsy harbored genotypic abnormalities that set coded by this gene is a cell signaling molecule that partici- them apart from malignant tissue. pates in every aspect of cellular physiologic and pathologic processes [33]. Recent studies [34, 35] suggest that an excess The16genes in the final modelspanlocialong 7chro- mosomes: 3p21: CTNNB1, 3q27: BCL6; 4q26: IL2; 6p21.3: PTP4A3 may play a key role in the acquisition of metastatic BAK1 and LTA; 8p12: FGFR, 8q24.12: MYC, 8q24.3: PTP4A3; potential of tumor cells. This study further supports gain 9p21: CDKN2A, CDKN2B; 11p13: LMO2, 11q13: CCND1, of PTP4A3 as a malignancy-associated alteration [36]in FGF3; 21q11.1: STCH, 21q22.3: TFF1. Gene alterations at HNSCC. these loci restate reported cytogenetic [7, 12–22] and mole- The MYC oncogene, located at 8q24.12, encodes a trans- cularlyaltered regionsbyLOH andarray CGHstudiesin cription factor that plays a key role in cell proliferation, diff- HNSCC [1, 5, 7, 23–28]. Additionally, copy number loss in erentiation, and apoptosis [37]. Gain of MYC was significant for laryngeal tumor progression [38], and the concomitant tumor and corresponding gain in nontumor, and vice versa, advocate aneuploidy events. The latter are highlighted for over expression of MYC and p53 oncogenes had worse dis- loss of CDKN2A (9p21), CTNNB1 (3p22), IL2 (4q26), LTA ease-free survival suggesting a role for p53 and MYC genes in progression of HNSCC [39]. In this study, gain of MYC (6p21.3), and TFF1 (21q22.3) in tumor, with corresponding gain in nontumor lesions, and gain of FGFR1 (8p12), MYC significantly discriminated tumor from nontumor tissue. Journal of Oncology 5 The corresponding loss of MYC copy number in nontumor progression during and after chemopreventive intervention suggests aneuploidy as a likely destabilizing event. [61, 62]. Though corresponding gain of CCND1 was not DNA double-strand breaks repair pathway has been im- retained in the final multivariate model, 24% of tumor plicated in maintaining genomic integrity via suppression of lesions had copy number gain. Loss of LMO2 (also known as chromosomal rearrangements. PRKDC (protein kinase, RBTN2 and TTG2), located at 11p13, predicted tumor le- DNA-activated, catalytic polypeptide) is associated with sions. This gene encodes a transcriptional cofactor critical for chromosomal instability with risk of breast and uterine cer- the development of hematopoietic stem cells [63]. vix cancer [40]. In this study, genomic instability at the 8q11 Gain of copy number at two 21q loci, TFF1 (21q22.3) locus favored gain of PRKDC copy number in tumor and and STCH (21q11.1), predicted nontumor lesions, and cor- corresponding loss in nontumor. responding loss of TFF1 was a significantly associated with FGFR-1, located at 8p21, had gain of copy number in tumor lesions. Loss of TFF1 in tumor lesions with corre- tumor and corresponding loss in nontumor. FGFR-1 expres- sponding gain in nontumor supports genomic instability as a sion hasbeendetectedinthyroid carcinoma [41]and in oral concerted early tumorigenesis event. STCH (stress 70 protein squamous cell carcinomas (OSCC). Amplification of FGFR1 chaperone), at 21q11.1, a member of the heat shock protein detected by FISH analysis on OSCC tissue microarray sec- 70 (HSP70) superfamily with cell-protective functions, was tions contributed to oral carcinogenesis at an early stage of previously identified as a candidate gene for susceptibility to development [42]. stomach cancer by genetic analyses [64]. STCH copy number Genetic alterations at the 9p21 locus have been linked to gain in nontumor lesions remained in the final model as an malignant progression in HNSCC [43, 44]. CDKN2A and independent predictor of nonmalignant tissue (correspond- CDKN2B genes map to 9p21 and are in tandem, spanning ing loss in tumor lesions remained a univariate variable, a region of approximately 80 kb [45, 46]. Inactivation of the P< 0.001). CDKN2A (p14), CDKN2B (p15), and CDKN2A (p16) genes The model’s discriminatory abilities (C-index/ROC of is a frequent event in human oral squamous cell carcinomas 0.93) support molecular distinctiveness of malignant versus INK4a [47]. The main modes of p16 inactivation in HNSCC nonmalignant tissue with significant predictive power. The are known to include homozygous deletions (40–60%), mu- latter is of particular significance because normal samples tations (15–20%), and gene hypermethylation events (5%) from patients with head and neck cancer, especially in the [47–49]. This study supports loss of CDKN2A and CDKN2B neighborhood of the tumor, can be genetically altered (field genes as independent predictors of tumor in HNSCC pa- cancerization). The proximity of tumor and nontumor le- tients. sions, therefore, makes it harder to discriminate between Amplification of the 11q13 amplicon is driven by multi- these two entities. However, the relatively large number of ple genes, rather than only one or two genes at this site [50– tissue records (n = 1,076) from 220 patients was a factor in 52]. In this study, four genes at the 11q13 locus were inter- overcoming these challenges to yield a robust model with rogated, CCND1, FGF3, EMS1,and RELA,ofwhich,gain excellent ability (C-index = 0.93) to discriminate malignant of CCND1, FGF3,and EMS1 were univariate (P< 0.0001) and nonmalignant tissue within the same patient. predictors of tumor. Multivariate analysis (P< 0.01) retained Genetic alterations at 16 chromosomal loci underscore gain of FGF3 in tumor and loss in nontumor, supporting in- the association of already known genes as well as newer gene volvement of amplification/gain of copy number of this gene targets in HNSCC pathogenesis. The sixteen gene predictors in HNSCC [53, 54]. FGF3 belongs to the basic fibroblast spanning loci along 7 chromosomes cover an array of essen- growth factor (FGF) gene family with a role in several impor- tial functions that ensure normal homeostasis to include tant cellular processes, including regulation of cell growth DNA repair (PRKDC), initiation of carcinogenesis (TFF1), and division, determination of cell type, formation of blood immune surveillance (IL2, LTA), cell cycle regulators vessels, wound healing, and embryo development. In (CDKN2A, CDKN2B), apoptosis (BAK1, STCH), regulation HNSCC, FGF3 had a significantly higher frequency of ampli- of cell proliferation and differentiation (CCND1, FGF3, ficationinhypopharyngealtumors[55]. Loss of CCND1 was MYC), transcription factors (BCL6), stem cell hematopoiesis significantly associated with nontumor lesions in this study (LMO2), adhesion, invasion and metastasis (CTNNB1, cohort, suggesting genomic instability/chromosomal ane- FGFR1), and acquisition of metastatic potential of tumor uploidy events in the direction of corresponding gain of cells (PTP4A3), implicating these genes as key players in the CCND1 in tumor lesions (P< 0.001, univariate analysis). tumorigenesis continuum. Overexpression and/or amplification of CCND1 is reported in 35%–65% of patients with HNSCC and is associated with poor prognosis [56–58]. Its expression is deregulated in pre- invasive lesions adjacent to invasive tumors and is associated 4. Conclusion with increased chromosomal instability and the likelihood of Genomic instability, a hallmark of malignant transforma- subsequent gene amplification. [59, 60]Lossof CCND1 copy tion, promotes a wide range of mutations, including chro- number in nontumor tissues in this study may reflect very mosome deletions, gene amplifications, translocations, and early genomic instability at this chromosomal locus and sup- polyploidy [40]. In this study, the directional loss and gain ports reports of CCND1 deregulation in preinvasive lesions of the upper aerodigestive tract with associated increased risk for several genes underscored the contribution of aneuploidy for the development of cancer accompanied by histologic in early HNSCC tumorigenesis. Our data support distinct 6 Journal of Oncology genetic signatures that discriminate malignant and nonma- confirms that microsatellite assay can identify patients at risk of developing oral squamous cell carcinoma within a field of lignant tissue in HNSCC. The 16 gene alteration signature in cancerization,” Cancer Research, vol. 60, no. 14, pp. 3893–3898, this study suggests finely choreographed genomic instability events to achieve biological distinctiveness, providing clues [12] T. E. Carey, C. J. Frank, J. R. Raval et al., “Identifying genetic to the drivers in invasive cancers as well as insight into gene changes associated with tumor progression in squamous cell rearrangements that might arise in nonmalignant lesions. carcinoma,” Acta Oto-Laryngologica, no. 529, pp. 229–232, Thegenesetsmeetstatistical rigortosegregate malignant squamous carcinoma lesions from nonmalignant lesions, [13] D. L. Buchhagen, M. J. Worsham, D. L. Van Dyke, and T. E. providing an opportunity for researchers to investigate these Carey, “Two regions of homozygosity on chromosome 3p in cancer-associated genes as potential targets of therapy either squamous cell carcinoma of the head and neck: comparison as single targets or as sets of targets when these occur in the with cytogenetic analysis,” Head and Neck,vol. 18, no.6,pp. same cancer lesion. 529–537, 1996. [14] T. E. Carey, M. J. Worsham, and D. L. Van Dyke, “Chromo- somal biomarkers in the clonal evolution of head and neck Acknowledgment squamous neoplasia,” Journal of Cellular Biochemistry, vol. 52, pp. 213–222, 1993. This research was supported by NIH R01 DE 15990 (M. J. [15] T. E. Carey, D. L. van Dyke, and M. J. Worsham, “Nonrandom Worsham). chromosome aberrations and clonal populations in head and neck cancer,” Anticancer Research, vol. 13, no. 6 B, pp. 2561– References 2567, 1993. [16] C. R. Bradford, K. A. Kimmel, D. L. Van Dyke et al., “11p [1] M. J. Worsham, G. Pals, J. P. Schouten et al., “Delineating deletions and breakpoints in squamous cell carcinoma: associ- genetic pathways of disease progression in head and neck ation with altered reactivity with the UM-E7 antibody,” Genes squamous cell carcinoma,” Archives of Otolaryngology-Head Chromosomes and Cancer, vol. 3, no. 4, pp. 272–282, 1991. and Neck Surgery, vol. 129, no. 7, pp. 702–708, 2003. [17] T. E. Carey, D. L. Van Dyke, M. J. Worsham et al., “Character- [2] J. P. Schouten, C. J. McElgunn, R. Waaijer, D. Zwijnenburg, ization of human laryngeal primary and metastatic squamous F. Diepvens, and G. Pals, “Relative quantification of 40 nu- cell carcinoma cell lines UM-SCC-17A and UM-SCC-17B,” cleic acid sequences by multiplex ligation-dependent probe Cancer Research, vol. 49, no. 21, pp. 6098–6107, 1989. amplification,” Nucleic Acids Research, vol. 30, no. 12, p. e57, [18] M. J. Worsham, M. J. Benninger, R. J. Zarbo, T. E. Carey, and D. L. Van Dyke, “Deletion 9p22-pter and loss of Y as primary [3] U. Raju, M. Lu, S. Sethi, H. Qureshi, S. R. Wolman, and M. chromosome abnormalities in a squamous cell carcinoma of J. Worsham, “Molecular classification of breast carcinoma in the vocal cord,” Genes Chromosomes and Cancer, vol. 6, no. 1, situ,” Current Genomics, vol. 7, no. 8, pp. 523–532, 2006. pp. 58–60, 1993. [4] J. K. Stephen, L. E. Vaught, M. C. Kang et al., “An epigenetically [19] M. J. Worsham, T. E. Carey, M. S. Benninger et al., “Clonal derived monoclonal origin for recurrent respiratory papillo- cytogenetic evolution in a squamous cell carcinoma of the skin matosis,” Archives of Otolaryngology-Head and Neck Surgery, from a xeroderma pigmentosum patient,” Genes Chromosomes vol. 133, no. 7, pp. 684–692, 2007. and Cancer, vol. 7, no. 3, pp. 158–164, 1993. [5] O. Saglam, V. Shah, and M. J. Worsham, “Molecular differen- [20] M. J. Worsham, S. R. Wolman,T.E.Carey,R.J.Zarbo,M. tiation of early and late stage laryngeal squamous cell carcino- S. Benninger, and D. L. Van Dyke, “Common clonal origin ma: an exploratory analysis,” Diagnostic Molecular Pathology, of synchronous primary head and neck squamous cell carci- vol. 16, no. 4, pp. 218–221, 2007. nomas: analysis by tumor karyotypes and fluorescence in situ [6] M. J. Worsham, G. Pals, J. P. Schouten et al., “High-resolution hybridization,” Human Pathology, vol. 26, no. 3, pp. 251–261, mapping of molecular events associated with immortalization, transformation, and progression to breast cancer in the [21] M. J. Worsham, S. R. Wolman,T.E.Carey,R.J.Zarbo,M. MCF10 model,” Breast Cancer Research and Treatment, vol. 96, S. Benninger, and D. L. Van Dyke, “Chromosomal aberrations no. 2, pp. 177–186, 2006. identified in culture of squamous carcinomas are confirmed by [7] M. J. Worsham, M. C. Kang, N. Tiwari et al., “Fine-mapping fluorescence in situ hybridisation,” Journal of Clinical Pathol- loss of gene architecture at the CDKN2B (p15INK4b), ogy, vol. 52, no. 1, pp. 42–46, 1999. CDKN2A (p14ARF, p16INK4a), and MTAP genes in head and [22] D. L. van Dyke, M. J. Worsham, M. S. Benninger et al., “Recur- neck squamous cell carcinoma,” Archives of Otolaryngology- rent cytogenetic abnormalities in squamous cell carcinomas of Head and Neck Surgery, vol. 132, no. 4, pp. 409–415, 2006. the head and neck region,” Genes Chromosomes and Cancer, [8] J. P. Kunjoonju, M. Raitanen, S. Grenman, ´ N. Tiwari, and M. vol. 9, no. 3, pp. 192–206, 1994. J. Worsham, “Identification of individual genes altered in [23] H. de Schutter,M.Spaepen,W.H.McBride,and S. Nuyts, squamous cell carcinoma of the vulva,” Genes Chromosomes “The clinical relevance of microsatellite alterations in head and and Cancer, vol. 44, no. 2, pp. 185–193, 2005. neck squamous cell carcinoma: a critical review,” European [9] A. Forastiere, W. Koch, A. Trotti, and D. Sidransky, “Head and Journal of Human Genetics, vol. 15, no. 7, pp. 734–741, 2007. neck cancer,” New England Journal of Medicine, vol. 345, no. [24] X. Zhou, R. C. K. Jordan, Y. Li, B. L. Huang, and D. T. W. 26, pp. 1890–1900, 2001. Wong, “Frequent allelic imbalances at 8p and 11q22 in oral [10] M. P. Rosin, X. Cheng, C. Poh et al., “Use of allelic loss to pre- and oropharyngeal epithelial dysplastic lesions,” Cancer Genet- dict malignant risk for low-grade oral epithelial dysplasia,” Cli- ics and Cytogenetics, vol. 161, no. 1, pp. 86–89, 2005. nical Cancer Research, vol. 6, no. 2, pp. 357–362, 2000. [25] S. W. Coon,A.T.Savera, R. J. Zarboetal., “Prognosticimpli- [11] M. Partridge, S. Pateromichelakis, E. Phillips, G. G. Emilion, R. P. A’Hern, and J. D. Langdon, “A case-control study cations of loss of heterozygosity at 8p21 and 9p21 in head and Journal of Oncology 7 neck squamous cell carcinoma,” International Journal of Can- in thyroid carcinoma,” Endocrinology, vol. 146, no. 3, pp. cer, vol. 111, no. 2, pp. 206–212, 2004. 1145–1153, 2005. [26] J. Akervall, “Genomic screening of head and neck cancer and [42] K. Freier, C. Schwaenen, C. Sticht et al., “Recurrent FGFR1 its implications for therapy planning,” European Archives of amplification and high FGFR1 protein expression in oral Oto-Rhino-Laryngology, vol. 263, no. 4, pp. 297–304, 2006. squamous cell carcinoma (OSCC),” Oral Oncology, vol. 43, no. [27] Q. Huang, G. P. Yu, S. A. McCormick et al., “Genetic diffe- 1, pp. 60–66, 2007. rences detected by comparative genomic hybridization in head [43] W. M. Lydiatt, B. J. Davidson, S. P. Schantz, S. Caruana, and and neck squamous cell carcinomas from different tumor sites: R. S. K. Chaganti, “9p21 deletion correlates with recurrence construction of oncogenetic trees for tumor progression,” in head and neck cancer,” Head and Neck,vol. 20, no.2,pp. Genes Chromosomes and Cancer, vol. 34, no. 2, pp. 224–233, 113–118, 1998. [44] A. L. Reed, J. Califano, P. Cairns et al., “High frequency of [28] K. Gotte, S. C. Tremmel, S. Popp et al., “Intratumoral genomic p16 (CDKN2/MTS-1/INK4A) inactivation in head and neck heterogeneity in advanced head and neck cancer detected by squamous cell carcinoma,” Cancer Research, vol. 56, no. 16, pp. comparative genomic hybridization,” Advances in Oto-Rhino- 3630–3633, 1996. Laryngology, vol. 62, pp. 38–48, 2005. [45] C. J. Sherr, “Cancer cell cycles,” Science, vol. 274, no. 5293, pp. [29] C. R. Leemans, B. J. M. Braakhuis, and R. H. Brakenhoff,“The 1672–1674, 1996. molecular biology of head and neck cancer,” Nature Reviews [46] C. J. Sherr, “The pezcoller lecture: cancer cell cycles revisited,” Cancer, vol. 11, no. 1, pp. 9–22, 2011. Cancer Research, vol. 60, no. 14, pp. 3689–3695, 2000. [30] S. Li, J. Jiao, Z. Lu, and M. Zhang, “An essential role for N- [47] S. Shintani, Y. Nakahara, M. Mihara, Y. Ueyama, and T. cadherin and β-catenin for progression in tongue squamous Matsumura, “Inactivation of the p14ARF, p15INK4B and cell carcinoma and their effect on invasion and metastasis of p16INK4A genes is a frequent event in human oral squamous Tca8113 tongue cancer cells,” Oncology Reports, vol. 21, no. 5, cell carcinomas,” Oral Oncology, vol. 37, no. 6, pp. 498–504, pp. 1223–1233, 2009. [31] M. V. Gonzalez,M.F.Pello,P.Ablanedo, C. Suar ´ ez, V. Alvarez, [48] E. P. Xing, Y. Nie, Y. Song et al., “Mechanisms of inactiva- and E. Coto, “Chromosome 3p loss of heterozygosity and mu- tion of p14(ARF), p15(INK4b), and p16(INK4a) genes in hu- tation analysis of the FHIT and β-cat genes in squamous cell man esophageal squamous cell carcinoma,” Clinical Cancer carcinoma of the head and neck,” Journal of Clinical Pathology, Research, vol. 5, no. 10, pp. 2704–2713, 1999. vol. 51, no. 7, pp. 520–524, 1998. [49] J. W. Rocco and D. Sidransky, “p16(MTS-1/CDKN2/INK4a) [32] X. Xie, O. P. Clausen, and M. Boysen, “Prognostic value of Bak in cancer progression,” Experimental Cell Research, vol. 264, expression in oral tongue squamous cell carcinomas,” Oncol- no. 1, pp. 42–55, 2001. ogy Reports, vol. 10, no. 2, pp. 369–374, 2003. [50] P. Gaudray, P. Szepetowski, C. Escot, D. Birnbaum, and C. [33] Q. Zeng,W.Hong, andY.H.Tan,“MousePRL-2 andPRL-3, Theillet, “DNA amplification at 11q13 in human cancer: from two potentially prenylated protein tyrosine phosphatases complexity to perplexity,” Mutation Research, vol. 276, no. 3, homologous to PRL-1,” Biochemical and Biophysical Research pp. 317–328, 1992. Communications, vol. 244, no. 2, pp. 421–427, 1998. [51] E. Schuuring, “The involvement of the chromosome 11q13 re- [34] F. Polato, A. Codegoni, R. Fruscio et al., “PRL-3 phosphatase is gion in human malignancies: cyclin D1 and EMS1 are two new implicated in ovarian cancer growth,” Clinical Cancer Re- candidate oncogenes-a review,” Gene, vol. 159, no. 1, pp. 83– search, vol. 11, no. 19, pp. 6835–6839, 2005. 96, 1995. [35] U. A. Miskad,S.Semba,H.Kato, andH.Yokozaki, “Expression [52] X. Huang, S. M. Gollin, S. Raja, and T. E. Godfrey, “High- of PRL-3 phosphatase in human gastric carcinomas: close resolution mapping of the 11q13 amplicon and identification correlation with invasion and metastasis,” Pathobiology, vol. of a gene, TAOS1, that is amplified and overexpressed in oral 71, no. 4, pp. 176–184, 2004. cancer cells,” Proceedings of the National Academy of Sciences of [36] M. Canel, P. Secades, M. Garzon-A ´ rango et al., “Involvement the United States of America, vol. 99, no. 17, pp. 11369–11374, of focal adhesion kinase in cellular invasion of head and neck squamous cell carcinomas via regulation of MMP-2 expres- [53] K. D. Somers, S. L. Cartwright, and G. L. Schechter, “Ampli- sion,” British Journal of Cancer, vol. 98, no. 7, pp. 1274–1284, fication of the int-2 gene in human head and neck squamous cell carcinomas,” Oncogene, vol. 5, no. 6, pp. 915–920, 1990. [37] C. V. Dang, “c-Myc target genes involved in cell growth, [54] J. S. Rubin, L. Qiu, and P. Etkind, “Amplification of the Int-2 apoptosis, and metabolism,” Molecular and Cellular Biology, gene in head and neck squamous cell carcinoma,” Journal of vol. 19, no. 1, pp. 1–11, 1999. Laryngology and Otology, vol. 109, no. 1, pp. 72–76, 1995. [38] D. K. Koynova, V. S. Tsenova, and D. I. Toncheva, “Tissue [55] J. P. Rodrigo, C. Suar ´ ez, M. V. Gonzalez ´ et al., “Variability of microarray analysis of C-MYC oncogene copy number genetic alterations in different sites of head and neck cancer,” changes in larynx carcinoma,” ORL, vol. 67, no. 2, pp. 92–95, Laryngoscope, vol. 111, no. 7, pp. 1297–1301, 2001. [56] J. A. Akervall, R. J. A. M. Michalides, H. Mineta et al., “Ampli- [39] A. F. L. Waitzberg, S. Nonogaki, I. N. Nishimoto et al., “Cli- nical significance of c-myc and p53 expression in head and fication of cyclin D1 in squamous cell carcinoma of the head and neck and the prognostic value of chromosomal abnormal- neck squamous cell carcinomas,” Cancer Detection and Preven- ities and cyclin D1 overexpression,” Cancer,vol. 79, no.2,pp. tion, vol. 28, no. 3, pp. 178–186, 2004. [40] M. Someya, K. I. Sakata, Y. Matsumoto et al., “The association 380–389, 1997. of DNA-dependent protein kinase activity with chromosomal [57] R. J. A. M. Michalides,N.M.J.Van Veelen,P.M.P.Kristel instability and risk of cancer,” Carcinogenesis,vol. 27, no.1,pp. et al., “Overexpression of cyclin D1 indicates a poor prognosis 117–122, 2006. in squamous cell carcinoma of the head and neck,” Archives [41] R. S. Bernard, L. Zheng, W. Liu, D. Winer, S. L. Asa, and S. of Otolaryngology-Head and Neck Surgery, vol. 123, no. 5, pp. Ezzat, “Fibroblast growth factor receptors as molecular targets 497–502, 1997. 8 Journal of Oncology [58] J. A. Akervall, Y. Jin, J. P. Wennerberg et al., “Chromosomal abnormalities involving 11q13 are associated with poor prog- nosis in patients with squamous cell carcinoma of the head and neck,” Cancer, vol. 76, no. 5, pp. 853–859, 1995. [59] J. G. Izzo, V. A. Papadimitrakopoulou, X. Q. Li et al., “Dys- regulated cyclin D1 expression early in head and neck tumori- genesis in vivo evidence for an association with subsequent gene amplification,” Oncogene, vol. 17, no. 18, pp. 2313–2322, [60] H. J. Roh, D. M. Shin, J. S. Lee et al., “Visualization of the tim- ing of gene amplification during multistep head and neck tu- morigenesis,” Cancer Research, vol. 60, no. 22, pp. 6496–6502, [61] D. L. Uhlman, G. Adams, D. Knapp, D. M. Aeppli, and G. Niehans, “Immunohistochemical staining for markers of fu- ture neoplastic progression in the larynx,” Cancer Research, vol. 56, no. 9, pp. 2199–2205, 1996. [62] V. A. Papadimitrakopoulou, J. Izzo, J. Keck et al., “Cyclin D1 and p16 alterations in advanced premalignant lesions of the upper aerodigestive tract: role in response to chemoprevention and cancer development,” Clinical Cancer Research, vol. 7, no. 10, pp. 3127–3134, 2001. [63] Y. Yamada, R. Pannell, A. Forster, and T. H. Rabbitts, “The oncogenic LIM-only transcription factor Lmo2 regulates an- giogenesis but not vasculogenesis in mice,” Proceedings of the National Academy of Sciences of the United States of America, vol. 97, no. 1, pp. 320–324, 2000. [64] N. Yamagata, K. Furuno, M. Sonoda, H. Sugimura, and K. Yamamoto, “Stomach cancer-derived del223V-226L mutation of the STCH gene causes loss of sensitization to TRAIL-med- iated apoptosis,” Biochemical and Biophysical Research Com- munications, vol. 376, no. 3, pp. 499–503, 2008. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Hindawi Publishing Corporation

Malignant and Nonmalignant Gene Signatures in Squamous Head and Neck Cancer

Loading next page...
 
/lp/hindawi-publishing-corporation/malignant-and-nonmalignant-gene-signatures-in-squamous-head-and-neck-ckbT8KTGm9

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2012 Maria J. Worsham et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1687-8450
eISSN
1687-8469
DOI
10.1155/2012/752860
Publisher site
See Article on Publisher Site

Abstract

Hindawi Publishing Corporation Journal of Oncology Volume 2012, Article ID 752860, 8 pages doi:10.1155/2012/752860 Research Article Malignant and Nonmalignant Gene Signatures in Squamous Head and Neck Cancer 1 2 1 1 Maria J. Worsham, Mei Lu, Kang Mei Chen, Josena K. Stephen, 1 1 Shaleta Havard, and Vanessa P. Schweitzer Department of Otolaryngology-Head & Neck Surgery, Henry Ford Health System, Detroit, MI 48202, USA Department of Biostatistics and Research Epidemiology, Henry Ford Health System, Detroit, MI 48202, USA Correspondence should be addressed to Maria J. Worsham, mworsha1@hfhs.org Received 7 December 2011; Revised 26 January 2012; Accepted 13 February 2012 Academic Editor: Mario A. Hermsen Copyright © 2012 Maria J. Worsham et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Genetic events specific to the pathogenesis of malignancy can offer clues to the tumorigenesis process. The objective of this study was to identify gene alterations that differentiate tumor and nontumor lesions in squamous head and neck cancer (HNSCC). DNA from 220 primary HNSCC with concurrently present tumor and nontumor lesions from the same patient was interrogated for ge- nomic alterations of loss or gain of copy. Conditional logistic regression dealt with tumor and non-tumor records within a patient. Of 113 genes, 53 had univariate effects (P< 0.01), of which 16 genes remained in the multivariable model withP< 0.01. The model had a C-index (ROC) of 0.93. Loss of CDKN2B and gain of BCL6, FGF3,and PTP4A3 predicted tumor. Loss of BAK1 and CCND1 and gain of STCH predicted nontumor. This highly powered model assigned alterations in 16 genes specific for malignant versus nonmalignant lesions, supporting their contribution to the pathogenesis of HNSCC as well as their potential utility as relevant tar- gets for further evaluation as markers of early detection and progression. 1. Introduction cancer genes in many tumor types; however, individual gene loci altered in tumors cannot be deduced solely from the type Knowledge of the genetic mechanisms that drive cancer of chromosome rearrangement [1]. Historically, the molecu- growth and development is important in understanding the lar pathogenesis of cancer has been teased out one gene at a pathogenesis of malignancy and provides insights into the time. Recent high-throughput genomewide candidate strat- tumorigenesis process. The underlying hypothesis is that be- egies such as the multiplex ligation-dependent probe ampli- havior of tumor cells is determined by genetic changes that fication (MLPA) assay [2] to identify specific genes for gain alter cell growth, cell differentiation, programmed cell death, and loss concurred with chromosomal aberrations and pro- and cell migration. Cancer is the result of transformation vide a novel index to estimate the extent of genomic abnor- from a normal to a malignant cell that results from accumu- mality with disease progression [1]. lated mutations. Acquisition of a fully malignant phenotype Molecular genetic prognosticators can influence preven- in colon cancer is thought to occur because of multiple steps tion, diagnosis, appropriateness of adjuvant chemotherapy, whose targets are alterations of growth-promoting onco- and, possibly, the chemotherapeutic regimen of cancer pa- genes and growth-inhibiting cancer suppressor genes. The tients. Dissecting out processes specific to the pathogenesis evolution in transformation from a normal squamous epi- of malignancy can distill key genetic biomarkers of HNSCC thelial cell to a cancer cell is likewise assumed to require sev- etiology, transformation, and progression. eral steps, some defined by genetic alteration. Genetic alterations provide means of identifying tumor In this study, in a primary HNSCC cohort of 220 patients, cells as well as defining changes that presumably determine with both tumor and nontumor lesions within a biopsy (tu- biological differences from their normal counterparts. Chro- mor and nontumor from the same patient), we examined mosome aberrations have served as landmarks to identify gene alterationsoflossand gain to derive multivariate 2 Journal of Oncology predictive models to discriminate malignant from nonma- reactions (probe sets p005, p006, p007, MRC Holland, Am- lignant lesions. sterdam). Briefly, DNA, diluted with water to a total volume of 5 μL, is denatured and fragmented by heating for 5 min- utes at 98 C in a thermocycler. Binary MLPA probes are add- 2. Methods ed and allowed to hybridize to their targets during an 16 hr incubation at 60 C, followed by the addition of dilution buf- 2.1. Patient Cohort. Cohort subjects were male and female fer and a ligase enzyme (MRC Holland, Amsterdam). During patients 18 years and older who underwent a HNSCC tissue the 15-minute incubation at 60 C, the two parts of a probe biopsy at the Henry Ford Health System from 1986–2006. become ligated to each other and become an amplifiable The use of formalin-fixed paraffin-embedded tissue blocks molecule if the complementary sequence is present in from patients with both tumor and nontumor records within the DNA sample. This is followed by the addition of PCR pri- the same biopsy and the collection of related patient infor- mers, dNTPs and Taq polymerase, followed by the following mation were approved by the Henry Ford Health System In- ◦ ◦ ◦ cycles: one at 1 min 95 C, 10 cycles: 30 sec 95 C, 30 sec 70 C, stitutional Review Board (IRB) Committee. ◦ ◦ ◦ ◦ 1min 72 C; 30 cycles at 30 sec 95 C, 30 sec 60 C, 1 min 72 C. In the event a study subject had more than one eligible Thesameprimerpair, oneofwhich is tagged with afluores- biopsy over the course of the study period, the primary bi- cent dye, amplifies all (ligated) probes. The relative amounts opsy was the index biopsy and the pathology report date marked the study enrollment (index) date. Cohort subjects of PCR product obtained reflect the relative amounts of ligat- missing biopsy tissue blocks or insufficient tissue for molec- ed probes at the start of the PCR reaction. Amplification pro- ular analysis were excluded. ducts are analyzed on a DNA sequencer (Applied Biosystems, Foster City, Ca), quantified and interpreted as previously des- cribed [1, 3, 4, 6–8]. 2.2. Histopathology. Pathology review of paraffin-embedded tissue sections captured all lesion types in a biopsy to include normal squamous epithelium, squamous dysplasia whether 2.4. Statistical Analysis. Conditional logistic regression mod- classified as mild, moderate, or severe/carcinoma in situ, and eling was used to address tumor and nontumor lesions with- tumor. Severe dysplasia and carcinoma in situ lesions were in the primary biopsy in an HNSCC patient. Analysis began groupedwithtumor andconsideredasmalignant outcomes. by testing individual genes as risk predictors/discriminators for tumor and nontumor (univariate analysis). Genes with 2.3. Molecular Analysis individual risks in a univariate analysis (P< 0.01) were can- didates for the first multivariable model. Prior to multivari- 2.3.1. Processing Lesion Specimens for Molecular Analysis. able modeling, genes were evaluated for their correlation and DNA was obtained from either whole 5 micron tissue sec- missing values. Highly correlated genes (correlation coeffi- tions (if the tissue block contained predominantly tumor or cient [r] > 0.7) or genes with larger missing values (>5%) nontumor tissue) or from microdissected tissue as previously were fitted separately along with other uncorrelated (r < 0.7) described [3]. Briefly, concurrently present tumor (severe genes. The stepwise model selection was considered. The dysplasia, carcinoma in situ, tumor) and nontumor (normal, final model included genes with P< 0.01 along with odds mild/moderate dysplasia) lesions in the same paraffin-em- ratios for loss or gain as risk predictors. The C-index/ROC bedded formalin fixed tissue block were marked by the path- (the receiver operating characteristic {ROC} curve), in a ologist and individual lesions were microdissected from 5 range of 0 to 1, is a measure of the model’s predictive ability, micron sections mounted on glass slides using a single-use where 0.5 indicates no discrimination and 0.7 or greater indi- disposable scalpel blade under a dissecting microscope. This procedure minimizes mixing of normal and tumor subpop- cates that the model is predictive. ulations and yields lesion and tumor samples estimated to be at least 90% free from contamination with normal cells 3. Results and Discussion [3, 4]. 3.1. Results. Matched tumor and nontumor lesions within 2.3.2. The Multiplex Ligation-Dependent Probe Amplification each patient in the 220 primary HNSCC cohort comprised (MLPA) Assay. MLPA has several advantages. It establishes a total of 1076 tissue records. There were 504 normal/mild/ the copy number of up to 41 nucleic acid sequences in one moderate dysplasia lesions (495: normal squamous epithe- single reaction. MLPA probes are able to discriminate bet- lium, 6: mild dysplasia, 3: moderate dysplasia), and 572 tu- ween sequences that differ in only one nucleotide. Moreover, mor lesions (568: tumor, 1: severe dysplasia, 3: carcinoma in MLPA reactions require a minimum of only 20 ng human situ (CIS)). Squamous mucosal dysplasia whether classified DNA making it especially amenable for PCR of DNA from as mild, moderate, or severe/carcinoma in situ is considered formalin-fixed paraffin-embedded head and neck tissues [3– intraepithelial neoplasia, and, as a precursor lesion group is 5]. separate from normal and tumor. The number of precursor, Starting with, approximately, 20–50 ng of genomic DNA, lesions is very small (13) in comparison to normal epithe- for each subject in the cohort, DNA from tumor and nontu- lium (495) and carcinomas (568). Of the 572 tumor lesions, mor is amplified for 122 probes (113 unique genes associated 193 (34%) were laryngeal, 170 (30%) were oral cavity, 151 with cancer including HNSCC) as separate sets of three (26%) were pharyngeal (oropharyngeal/hypopharyngeal), Journal of Oncology 3 Table 1: Genes alterations that predict malignant (M, bolded) and nonmalignant (NM, italicized). Effect Chromosome Odds ratio estimate Lower CL Upper CL §M CTNNB1 lossversusnormal 3p22 2.682 1.394 5.162 NM CTNNB1 gain versus normal 0.323 0.147 0.71 BCL6 loss versus normal 3q27 0.55 0.27 1.12 BCL6 gain versus normal 8.989 3.155 25.612 §M IL2 loss versus normal 4q26 3.697 1.774 7.705 NM IL2 gain versus normal 0.149 0.055 0.407 NM BAK1 lossversusnormal 6p21.3 0.262 0.103 0.666 BAK1 gain versus normal 0.438 0.192 0.999 §M LTA lossversusnormal 6p21.3 2.156 1.172 3.965 NM LTA gain versus normal 0.266 0.108 0.655 §NM FGFR1 lossversusnormal 8p21 0.275 0.126 0.598 FGFR1 gain versus normal 5.555 1.689 18.267 §NM PRKDC loss versus normal 8q11 0.276 0.11 0.692 PRKDC gain versus normal 5.449 2.09 14.206 §NM MYC loss versus normal 8q24.12 0.221 0.097 0.503 MYC gain versus normal 2.218 1.136 4.332 PTP4A3 loss versus normal 8q24.3 0.493 0.214 1.135 PTP4A3 gain versus normal 12.158 3.461 42.71 §M CDKN2A loss versus normal 9p21 1.845 1.013 3.362 NM CDKN2A gain versus normal 0.14 0.056 0.35 CDKN2B loss versus normal 9p21 3.256 1.676 6.325 CDKN2B gain versus normal 1.168 0.442 3.087 LMO2 loss versus normal 11p13 4.977 2.16 11.466 LMO2 gain versus normal 0.573 0.205 1.607 FGF3 loss versus normal 11q13 0.882 0.447 1.741 FGF3 gain versus normal 7.819 3.286 18.604 NM CCND1 loss versus normal 11q13 0.403 0.22 0.736 CCND1 gain versus normal 1.239 0.634 2.421 STCH loss versus normal 21q11.1 1.788 0.833 3.839 NM STCH gain versus normal 0.124 0.043 0.359 §M TFF1 loss versus normal 21q22.3 3.019 1.514 6.02 NM TFF1 gain versus normal 0.08 0.024 0.268 CL: confidence limit, : genes with loss and gain signifying aneuploidy. and 58 (10%) were lesions in the other category (nasophar- patients, 25% (55/220) were female and 75% (165/220) were ynx, nasal cavity, paranasal sinuses). male. The 1076 lesions were distributed among 932 tissue The missing value for each gene was in a range of 0% to blocks, of which 434 (47%) had tumor lesions only, 363 4.2%. Of the 113 unique genes (selected based on their asso- (39%) had nontumor lesions only, and 135 (14%) tissue ciation with cancer including HNSCC), 53 genes had uni- blocks had both tumor and nontumor lesions. Within a pa- variate effects (P< 0.01) and were considered as the candi- tient, the number of nontumor lesions ranged from 1 to 6 date genes for multivariable analyses. After the stepwise and from 1 to 7 for tumor lesions. Of the 220 patients with model selection, 16 genes remained in the multivariable both tumor and nontumor within the same patient, 35 model (P< 0.01) (Table 1). The model had a C-index (re- (16%) had 1 tumor and nontumor record, 28 (13%) had 2 ceiver operating characteristic (ROC)) of 0.93. tumor and nontumor records, 27 (12%) had 3 tumor and The 16 genes in the final model with alterations of loss nontumor records. and/or gain accounted for loci along 7 chromosomes: 3, 4, Fifty percent of cohort (110/220) subjects were Caucasian 6, 8, 9, 11, and 21 (Table 1). Of these, 50% were altered in American (CA), 38% (84/220) were self-reported as African both tumor and nontumor, with loss or gain reflective of American (AA), 4% (9/220) were non-CA/non-AA (His- chromosomal aneuploidy. This copy number instability panic: 2, Asian Pacific Islander: 2, Middle Eastern: 3, other: favored loss of CDKN2A (9p21), CTNNB1 (3p21), IL2 2), and, in 8% (17/220), race was missing. Of the 220 (4q26), LTA (6p21.3), and TFF1 (21q22.3) in tumor, with 4 Journal of Oncology corresponding gain in nontumor lesions, and gain of FGFR1 (8q24.12), and PRKDC (8q11) in tumor, with corresponding (8p21), MYC (8q24.12), and PRKDC (8q11) in tumor, with loss in nontumor. Chromosomal instability occurs early corresponding loss in nontumor (Table 1). along the tumorigenesis continuum and aneuploidy at the Loss of CDKN2B (9p21) and LMO2 (11p13) and gain of 9p21 locus affecting corresponding loss and gain in tumor BCL6 (3q27), FGF3 (11q13), and PTP4A3 (8q24.3) predicted and normal tissue, respectively, concurs with the proposed tumor. Loss of BAK1 (6p21.3) and CCND1 (11q13) and gain postulated model of molecular carcinogenesis for HNSCC of STCH (21q11.1) predicted nontumor (Table 1). [29]. In HNSCC, chromosomal aberrations on the long arm of Analysis excluding the 6 mild dysplasia and 3 moderate chromosome 3, resulting in gain of distal 3q segments, have dysplasia lesions from the nontumor group and the 1 severe been reported as recurring karyotypic alterations [22]. Gain dysplasia lesion and 3 CIS from the tumor group generated of 3q is supported by increased copy number (3-4 copies) of an identical multivariable model outcome. PIK3CA at 3q26.3, MME (3q25.1), and BCL6 genes at 3q27 [1] in HNSCC. In this study, gain of BCL6 was significantly 3.2. Discussion. Cancerous tissue in most cases has a dis- associated with tumor lesions. The protein encoded by BCL6 tinctive appearance under the microscope. Distinguishing is a zinc finger transcription factor and acts as a sequence- traits include a large number of dividing cells, variation in specific repressor of transcription. nuclear size and shape, variation in cell size and shape, loss Another chromosome 3 gene, CTNNB1 (catenin beta-1) of specialized cell features, loss of normal tissue organization, in the short arm at 3p21, is an adherens junction protein, and a poorly defined tumor boundary. Biopsy and micro- closely associated with adhesion, invasion, and metastasis in scopical examination can also distinguish malignancy, pre- different types of tumors, including SCC of the tongue [30]. cursor lesions of carcinoma in situ, mild, moderate, and The 3p21 region had the highest rate of allelic deletion (63%) severe dysplasia, and less reliably, hyperplasia, from normal in HNSCC [31]and is supportedbylossof CTNNB1 in tu- appearing tissue. mor lesions in this study. Corresponding gain of CTNNB1 In the multistep process of tumorigenesis, hyperplasia, copy number in nontumor lesions underscores chromoso- which refers to tissue growth based on an excessive rate of cell mal instability and ensuing aneuploidy as early events in the division, leading to a larger than usual number of cells but tumorigenesis process. with a normal orderly arrangement of cells within the tissue, Loss and corresponding gain of IL2 at 4q26 was signifi- and considered reversible, is thought to precede dysplasia. cantly associated with tumor and nontumor lesions, respec- Dysplasia, an abnormal type of excessive cell proliferation tively. The IL2 protein is produced by T cells in response to characterized by loss of normal tissue arrangement and cell antigenic or mitogenic stimulation and is required for T-cell structure, may revert to normal behavior, but, occasionally, proliferation and other activities crucial to regulation of the these lesions gradually become malignant. Distinguishing immune response. true precursor lesions on the basis of morphology alone is BAK1 (6p21.3) is a proapoptotic member of the BCL-2 often unreliable. In the molecular progression of HNSCC, family of genes that are involved in regulation of program- normal or minimal dysplasia often harbors abnormal geno- med cell death, and its increased expression had poorer dis- ease-specific survival in oral tongue squamous cell carcino- types [9], which do not necessarily correlate with observable changes in phenotype (morphology) [10, 11]. mas [32]. As a corollary to increased expression [32], in this In this study, 16 gene alterations with significant discrim- study, loss of BAK1 was a nontumor-specific event. Gene alterations were noted for 4 genes on chromosome inatory ability differentiated malignant HNSCC from non- malignant tissue. For matched tumor and nontumor lesions 8, three on 8q and one on 8p. Gains or amplifications involv- from the same patient within the 220 primary HNSCC study ing chromosome arm 8q are one of the most recurrent chro- cohort, examined for alterations in 113 unique genes with mosomal alterations in head and neck tumors. The human association to head and neck cancer, molecular alterations in protein tyrosine phosphatase type IVA, member 3, also known as PTP4A3, is located at 8q24.3 [33]. The protein en- “normal” appearing epithelium within the environment of a malignant biopsy harbored genotypic abnormalities that set coded by this gene is a cell signaling molecule that partici- them apart from malignant tissue. pates in every aspect of cellular physiologic and pathologic processes [33]. Recent studies [34, 35] suggest that an excess The16genes in the final modelspanlocialong 7chro- mosomes: 3p21: CTNNB1, 3q27: BCL6; 4q26: IL2; 6p21.3: PTP4A3 may play a key role in the acquisition of metastatic BAK1 and LTA; 8p12: FGFR, 8q24.12: MYC, 8q24.3: PTP4A3; potential of tumor cells. This study further supports gain 9p21: CDKN2A, CDKN2B; 11p13: LMO2, 11q13: CCND1, of PTP4A3 as a malignancy-associated alteration [36]in FGF3; 21q11.1: STCH, 21q22.3: TFF1. Gene alterations at HNSCC. these loci restate reported cytogenetic [7, 12–22] and mole- The MYC oncogene, located at 8q24.12, encodes a trans- cularlyaltered regionsbyLOH andarray CGHstudiesin cription factor that plays a key role in cell proliferation, diff- HNSCC [1, 5, 7, 23–28]. Additionally, copy number loss in erentiation, and apoptosis [37]. Gain of MYC was significant for laryngeal tumor progression [38], and the concomitant tumor and corresponding gain in nontumor, and vice versa, advocate aneuploidy events. The latter are highlighted for over expression of MYC and p53 oncogenes had worse dis- loss of CDKN2A (9p21), CTNNB1 (3p22), IL2 (4q26), LTA ease-free survival suggesting a role for p53 and MYC genes in progression of HNSCC [39]. In this study, gain of MYC (6p21.3), and TFF1 (21q22.3) in tumor, with corresponding gain in nontumor lesions, and gain of FGFR1 (8p12), MYC significantly discriminated tumor from nontumor tissue. Journal of Oncology 5 The corresponding loss of MYC copy number in nontumor progression during and after chemopreventive intervention suggests aneuploidy as a likely destabilizing event. [61, 62]. Though corresponding gain of CCND1 was not DNA double-strand breaks repair pathway has been im- retained in the final multivariate model, 24% of tumor plicated in maintaining genomic integrity via suppression of lesions had copy number gain. Loss of LMO2 (also known as chromosomal rearrangements. PRKDC (protein kinase, RBTN2 and TTG2), located at 11p13, predicted tumor le- DNA-activated, catalytic polypeptide) is associated with sions. This gene encodes a transcriptional cofactor critical for chromosomal instability with risk of breast and uterine cer- the development of hematopoietic stem cells [63]. vix cancer [40]. In this study, genomic instability at the 8q11 Gain of copy number at two 21q loci, TFF1 (21q22.3) locus favored gain of PRKDC copy number in tumor and and STCH (21q11.1), predicted nontumor lesions, and cor- corresponding loss in nontumor. responding loss of TFF1 was a significantly associated with FGFR-1, located at 8p21, had gain of copy number in tumor lesions. Loss of TFF1 in tumor lesions with corre- tumor and corresponding loss in nontumor. FGFR-1 expres- sponding gain in nontumor supports genomic instability as a sion hasbeendetectedinthyroid carcinoma [41]and in oral concerted early tumorigenesis event. STCH (stress 70 protein squamous cell carcinomas (OSCC). Amplification of FGFR1 chaperone), at 21q11.1, a member of the heat shock protein detected by FISH analysis on OSCC tissue microarray sec- 70 (HSP70) superfamily with cell-protective functions, was tions contributed to oral carcinogenesis at an early stage of previously identified as a candidate gene for susceptibility to development [42]. stomach cancer by genetic analyses [64]. STCH copy number Genetic alterations at the 9p21 locus have been linked to gain in nontumor lesions remained in the final model as an malignant progression in HNSCC [43, 44]. CDKN2A and independent predictor of nonmalignant tissue (correspond- CDKN2B genes map to 9p21 and are in tandem, spanning ing loss in tumor lesions remained a univariate variable, a region of approximately 80 kb [45, 46]. Inactivation of the P< 0.001). CDKN2A (p14), CDKN2B (p15), and CDKN2A (p16) genes The model’s discriminatory abilities (C-index/ROC of is a frequent event in human oral squamous cell carcinomas 0.93) support molecular distinctiveness of malignant versus INK4a [47]. The main modes of p16 inactivation in HNSCC nonmalignant tissue with significant predictive power. The are known to include homozygous deletions (40–60%), mu- latter is of particular significance because normal samples tations (15–20%), and gene hypermethylation events (5%) from patients with head and neck cancer, especially in the [47–49]. This study supports loss of CDKN2A and CDKN2B neighborhood of the tumor, can be genetically altered (field genes as independent predictors of tumor in HNSCC pa- cancerization). The proximity of tumor and nontumor le- tients. sions, therefore, makes it harder to discriminate between Amplification of the 11q13 amplicon is driven by multi- these two entities. However, the relatively large number of ple genes, rather than only one or two genes at this site [50– tissue records (n = 1,076) from 220 patients was a factor in 52]. In this study, four genes at the 11q13 locus were inter- overcoming these challenges to yield a robust model with rogated, CCND1, FGF3, EMS1,and RELA,ofwhich,gain excellent ability (C-index = 0.93) to discriminate malignant of CCND1, FGF3,and EMS1 were univariate (P< 0.0001) and nonmalignant tissue within the same patient. predictors of tumor. Multivariate analysis (P< 0.01) retained Genetic alterations at 16 chromosomal loci underscore gain of FGF3 in tumor and loss in nontumor, supporting in- the association of already known genes as well as newer gene volvement of amplification/gain of copy number of this gene targets in HNSCC pathogenesis. The sixteen gene predictors in HNSCC [53, 54]. FGF3 belongs to the basic fibroblast spanning loci along 7 chromosomes cover an array of essen- growth factor (FGF) gene family with a role in several impor- tial functions that ensure normal homeostasis to include tant cellular processes, including regulation of cell growth DNA repair (PRKDC), initiation of carcinogenesis (TFF1), and division, determination of cell type, formation of blood immune surveillance (IL2, LTA), cell cycle regulators vessels, wound healing, and embryo development. In (CDKN2A, CDKN2B), apoptosis (BAK1, STCH), regulation HNSCC, FGF3 had a significantly higher frequency of ampli- of cell proliferation and differentiation (CCND1, FGF3, ficationinhypopharyngealtumors[55]. Loss of CCND1 was MYC), transcription factors (BCL6), stem cell hematopoiesis significantly associated with nontumor lesions in this study (LMO2), adhesion, invasion and metastasis (CTNNB1, cohort, suggesting genomic instability/chromosomal ane- FGFR1), and acquisition of metastatic potential of tumor uploidy events in the direction of corresponding gain of cells (PTP4A3), implicating these genes as key players in the CCND1 in tumor lesions (P< 0.001, univariate analysis). tumorigenesis continuum. Overexpression and/or amplification of CCND1 is reported in 35%–65% of patients with HNSCC and is associated with poor prognosis [56–58]. Its expression is deregulated in pre- invasive lesions adjacent to invasive tumors and is associated 4. Conclusion with increased chromosomal instability and the likelihood of Genomic instability, a hallmark of malignant transforma- subsequent gene amplification. [59, 60]Lossof CCND1 copy tion, promotes a wide range of mutations, including chro- number in nontumor tissues in this study may reflect very mosome deletions, gene amplifications, translocations, and early genomic instability at this chromosomal locus and sup- polyploidy [40]. In this study, the directional loss and gain ports reports of CCND1 deregulation in preinvasive lesions of the upper aerodigestive tract with associated increased risk for several genes underscored the contribution of aneuploidy for the development of cancer accompanied by histologic in early HNSCC tumorigenesis. Our data support distinct 6 Journal of Oncology genetic signatures that discriminate malignant and nonma- confirms that microsatellite assay can identify patients at risk of developing oral squamous cell carcinoma within a field of lignant tissue in HNSCC. The 16 gene alteration signature in cancerization,” Cancer Research, vol. 60, no. 14, pp. 3893–3898, this study suggests finely choreographed genomic instability events to achieve biological distinctiveness, providing clues [12] T. E. Carey, C. J. Frank, J. R. Raval et al., “Identifying genetic to the drivers in invasive cancers as well as insight into gene changes associated with tumor progression in squamous cell rearrangements that might arise in nonmalignant lesions. carcinoma,” Acta Oto-Laryngologica, no. 529, pp. 229–232, Thegenesetsmeetstatistical rigortosegregate malignant squamous carcinoma lesions from nonmalignant lesions, [13] D. L. Buchhagen, M. J. Worsham, D. L. Van Dyke, and T. E. providing an opportunity for researchers to investigate these Carey, “Two regions of homozygosity on chromosome 3p in cancer-associated genes as potential targets of therapy either squamous cell carcinoma of the head and neck: comparison as single targets or as sets of targets when these occur in the with cytogenetic analysis,” Head and Neck,vol. 18, no.6,pp. same cancer lesion. 529–537, 1996. [14] T. E. Carey, M. J. Worsham, and D. L. Van Dyke, “Chromo- somal biomarkers in the clonal evolution of head and neck Acknowledgment squamous neoplasia,” Journal of Cellular Biochemistry, vol. 52, pp. 213–222, 1993. This research was supported by NIH R01 DE 15990 (M. J. [15] T. E. Carey, D. L. van Dyke, and M. J. Worsham, “Nonrandom Worsham). chromosome aberrations and clonal populations in head and neck cancer,” Anticancer Research, vol. 13, no. 6 B, pp. 2561– References 2567, 1993. [16] C. R. Bradford, K. A. Kimmel, D. L. Van Dyke et al., “11p [1] M. J. Worsham, G. Pals, J. P. Schouten et al., “Delineating deletions and breakpoints in squamous cell carcinoma: associ- genetic pathways of disease progression in head and neck ation with altered reactivity with the UM-E7 antibody,” Genes squamous cell carcinoma,” Archives of Otolaryngology-Head Chromosomes and Cancer, vol. 3, no. 4, pp. 272–282, 1991. and Neck Surgery, vol. 129, no. 7, pp. 702–708, 2003. [17] T. E. Carey, D. L. Van Dyke, M. J. Worsham et al., “Character- [2] J. P. Schouten, C. J. McElgunn, R. Waaijer, D. Zwijnenburg, ization of human laryngeal primary and metastatic squamous F. Diepvens, and G. Pals, “Relative quantification of 40 nu- cell carcinoma cell lines UM-SCC-17A and UM-SCC-17B,” cleic acid sequences by multiplex ligation-dependent probe Cancer Research, vol. 49, no. 21, pp. 6098–6107, 1989. amplification,” Nucleic Acids Research, vol. 30, no. 12, p. e57, [18] M. J. Worsham, M. J. Benninger, R. J. Zarbo, T. E. Carey, and D. L. Van Dyke, “Deletion 9p22-pter and loss of Y as primary [3] U. Raju, M. Lu, S. Sethi, H. Qureshi, S. R. Wolman, and M. chromosome abnormalities in a squamous cell carcinoma of J. Worsham, “Molecular classification of breast carcinoma in the vocal cord,” Genes Chromosomes and Cancer, vol. 6, no. 1, situ,” Current Genomics, vol. 7, no. 8, pp. 523–532, 2006. pp. 58–60, 1993. [4] J. K. Stephen, L. E. Vaught, M. C. Kang et al., “An epigenetically [19] M. J. Worsham, T. E. Carey, M. S. Benninger et al., “Clonal derived monoclonal origin for recurrent respiratory papillo- cytogenetic evolution in a squamous cell carcinoma of the skin matosis,” Archives of Otolaryngology-Head and Neck Surgery, from a xeroderma pigmentosum patient,” Genes Chromosomes vol. 133, no. 7, pp. 684–692, 2007. and Cancer, vol. 7, no. 3, pp. 158–164, 1993. [5] O. Saglam, V. Shah, and M. J. Worsham, “Molecular differen- [20] M. J. Worsham, S. R. Wolman,T.E.Carey,R.J.Zarbo,M. tiation of early and late stage laryngeal squamous cell carcino- S. Benninger, and D. L. Van Dyke, “Common clonal origin ma: an exploratory analysis,” Diagnostic Molecular Pathology, of synchronous primary head and neck squamous cell carci- vol. 16, no. 4, pp. 218–221, 2007. nomas: analysis by tumor karyotypes and fluorescence in situ [6] M. J. Worsham, G. Pals, J. P. Schouten et al., “High-resolution hybridization,” Human Pathology, vol. 26, no. 3, pp. 251–261, mapping of molecular events associated with immortalization, transformation, and progression to breast cancer in the [21] M. J. Worsham, S. R. Wolman,T.E.Carey,R.J.Zarbo,M. MCF10 model,” Breast Cancer Research and Treatment, vol. 96, S. Benninger, and D. L. Van Dyke, “Chromosomal aberrations no. 2, pp. 177–186, 2006. identified in culture of squamous carcinomas are confirmed by [7] M. J. Worsham, M. C. Kang, N. Tiwari et al., “Fine-mapping fluorescence in situ hybridisation,” Journal of Clinical Pathol- loss of gene architecture at the CDKN2B (p15INK4b), ogy, vol. 52, no. 1, pp. 42–46, 1999. CDKN2A (p14ARF, p16INK4a), and MTAP genes in head and [22] D. L. van Dyke, M. J. Worsham, M. S. Benninger et al., “Recur- neck squamous cell carcinoma,” Archives of Otolaryngology- rent cytogenetic abnormalities in squamous cell carcinomas of Head and Neck Surgery, vol. 132, no. 4, pp. 409–415, 2006. the head and neck region,” Genes Chromosomes and Cancer, [8] J. P. Kunjoonju, M. Raitanen, S. Grenman, ´ N. Tiwari, and M. vol. 9, no. 3, pp. 192–206, 1994. J. Worsham, “Identification of individual genes altered in [23] H. de Schutter,M.Spaepen,W.H.McBride,and S. Nuyts, squamous cell carcinoma of the vulva,” Genes Chromosomes “The clinical relevance of microsatellite alterations in head and and Cancer, vol. 44, no. 2, pp. 185–193, 2005. neck squamous cell carcinoma: a critical review,” European [9] A. Forastiere, W. Koch, A. Trotti, and D. Sidransky, “Head and Journal of Human Genetics, vol. 15, no. 7, pp. 734–741, 2007. neck cancer,” New England Journal of Medicine, vol. 345, no. [24] X. Zhou, R. C. K. Jordan, Y. Li, B. L. Huang, and D. T. W. 26, pp. 1890–1900, 2001. Wong, “Frequent allelic imbalances at 8p and 11q22 in oral [10] M. P. Rosin, X. Cheng, C. Poh et al., “Use of allelic loss to pre- and oropharyngeal epithelial dysplastic lesions,” Cancer Genet- dict malignant risk for low-grade oral epithelial dysplasia,” Cli- ics and Cytogenetics, vol. 161, no. 1, pp. 86–89, 2005. nical Cancer Research, vol. 6, no. 2, pp. 357–362, 2000. [25] S. W. Coon,A.T.Savera, R. J. Zarboetal., “Prognosticimpli- [11] M. Partridge, S. Pateromichelakis, E. Phillips, G. G. Emilion, R. P. A’Hern, and J. D. Langdon, “A case-control study cations of loss of heterozygosity at 8p21 and 9p21 in head and Journal of Oncology 7 neck squamous cell carcinoma,” International Journal of Can- in thyroid carcinoma,” Endocrinology, vol. 146, no. 3, pp. cer, vol. 111, no. 2, pp. 206–212, 2004. 1145–1153, 2005. [26] J. Akervall, “Genomic screening of head and neck cancer and [42] K. Freier, C. Schwaenen, C. Sticht et al., “Recurrent FGFR1 its implications for therapy planning,” European Archives of amplification and high FGFR1 protein expression in oral Oto-Rhino-Laryngology, vol. 263, no. 4, pp. 297–304, 2006. squamous cell carcinoma (OSCC),” Oral Oncology, vol. 43, no. [27] Q. Huang, G. P. Yu, S. A. McCormick et al., “Genetic diffe- 1, pp. 60–66, 2007. rences detected by comparative genomic hybridization in head [43] W. M. Lydiatt, B. J. Davidson, S. P. Schantz, S. Caruana, and and neck squamous cell carcinomas from different tumor sites: R. S. K. Chaganti, “9p21 deletion correlates with recurrence construction of oncogenetic trees for tumor progression,” in head and neck cancer,” Head and Neck,vol. 20, no.2,pp. Genes Chromosomes and Cancer, vol. 34, no. 2, pp. 224–233, 113–118, 1998. [44] A. L. Reed, J. Califano, P. Cairns et al., “High frequency of [28] K. Gotte, S. C. Tremmel, S. Popp et al., “Intratumoral genomic p16 (CDKN2/MTS-1/INK4A) inactivation in head and neck heterogeneity in advanced head and neck cancer detected by squamous cell carcinoma,” Cancer Research, vol. 56, no. 16, pp. comparative genomic hybridization,” Advances in Oto-Rhino- 3630–3633, 1996. Laryngology, vol. 62, pp. 38–48, 2005. [45] C. J. Sherr, “Cancer cell cycles,” Science, vol. 274, no. 5293, pp. [29] C. R. Leemans, B. J. M. Braakhuis, and R. H. Brakenhoff,“The 1672–1674, 1996. molecular biology of head and neck cancer,” Nature Reviews [46] C. J. Sherr, “The pezcoller lecture: cancer cell cycles revisited,” Cancer, vol. 11, no. 1, pp. 9–22, 2011. Cancer Research, vol. 60, no. 14, pp. 3689–3695, 2000. [30] S. Li, J. Jiao, Z. Lu, and M. Zhang, “An essential role for N- [47] S. Shintani, Y. Nakahara, M. Mihara, Y. Ueyama, and T. cadherin and β-catenin for progression in tongue squamous Matsumura, “Inactivation of the p14ARF, p15INK4B and cell carcinoma and their effect on invasion and metastasis of p16INK4A genes is a frequent event in human oral squamous Tca8113 tongue cancer cells,” Oncology Reports, vol. 21, no. 5, cell carcinomas,” Oral Oncology, vol. 37, no. 6, pp. 498–504, pp. 1223–1233, 2009. [31] M. V. Gonzalez,M.F.Pello,P.Ablanedo, C. Suar ´ ez, V. Alvarez, [48] E. P. Xing, Y. Nie, Y. Song et al., “Mechanisms of inactiva- and E. Coto, “Chromosome 3p loss of heterozygosity and mu- tion of p14(ARF), p15(INK4b), and p16(INK4a) genes in hu- tation analysis of the FHIT and β-cat genes in squamous cell man esophageal squamous cell carcinoma,” Clinical Cancer carcinoma of the head and neck,” Journal of Clinical Pathology, Research, vol. 5, no. 10, pp. 2704–2713, 1999. vol. 51, no. 7, pp. 520–524, 1998. [49] J. W. Rocco and D. Sidransky, “p16(MTS-1/CDKN2/INK4a) [32] X. Xie, O. P. Clausen, and M. Boysen, “Prognostic value of Bak in cancer progression,” Experimental Cell Research, vol. 264, expression in oral tongue squamous cell carcinomas,” Oncol- no. 1, pp. 42–55, 2001. ogy Reports, vol. 10, no. 2, pp. 369–374, 2003. [50] P. Gaudray, P. Szepetowski, C. Escot, D. Birnbaum, and C. [33] Q. Zeng,W.Hong, andY.H.Tan,“MousePRL-2 andPRL-3, Theillet, “DNA amplification at 11q13 in human cancer: from two potentially prenylated protein tyrosine phosphatases complexity to perplexity,” Mutation Research, vol. 276, no. 3, homologous to PRL-1,” Biochemical and Biophysical Research pp. 317–328, 1992. Communications, vol. 244, no. 2, pp. 421–427, 1998. [51] E. Schuuring, “The involvement of the chromosome 11q13 re- [34] F. Polato, A. Codegoni, R. Fruscio et al., “PRL-3 phosphatase is gion in human malignancies: cyclin D1 and EMS1 are two new implicated in ovarian cancer growth,” Clinical Cancer Re- candidate oncogenes-a review,” Gene, vol. 159, no. 1, pp. 83– search, vol. 11, no. 19, pp. 6835–6839, 2005. 96, 1995. [35] U. A. Miskad,S.Semba,H.Kato, andH.Yokozaki, “Expression [52] X. Huang, S. M. Gollin, S. Raja, and T. E. Godfrey, “High- of PRL-3 phosphatase in human gastric carcinomas: close resolution mapping of the 11q13 amplicon and identification correlation with invasion and metastasis,” Pathobiology, vol. of a gene, TAOS1, that is amplified and overexpressed in oral 71, no. 4, pp. 176–184, 2004. cancer cells,” Proceedings of the National Academy of Sciences of [36] M. Canel, P. Secades, M. Garzon-A ´ rango et al., “Involvement the United States of America, vol. 99, no. 17, pp. 11369–11374, of focal adhesion kinase in cellular invasion of head and neck squamous cell carcinomas via regulation of MMP-2 expres- [53] K. D. Somers, S. L. Cartwright, and G. L. Schechter, “Ampli- sion,” British Journal of Cancer, vol. 98, no. 7, pp. 1274–1284, fication of the int-2 gene in human head and neck squamous cell carcinomas,” Oncogene, vol. 5, no. 6, pp. 915–920, 1990. [37] C. V. Dang, “c-Myc target genes involved in cell growth, [54] J. S. Rubin, L. Qiu, and P. Etkind, “Amplification of the Int-2 apoptosis, and metabolism,” Molecular and Cellular Biology, gene in head and neck squamous cell carcinoma,” Journal of vol. 19, no. 1, pp. 1–11, 1999. Laryngology and Otology, vol. 109, no. 1, pp. 72–76, 1995. [38] D. K. Koynova, V. S. Tsenova, and D. I. Toncheva, “Tissue [55] J. P. Rodrigo, C. Suar ´ ez, M. V. Gonzalez ´ et al., “Variability of microarray analysis of C-MYC oncogene copy number genetic alterations in different sites of head and neck cancer,” changes in larynx carcinoma,” ORL, vol. 67, no. 2, pp. 92–95, Laryngoscope, vol. 111, no. 7, pp. 1297–1301, 2001. [56] J. A. Akervall, R. J. A. M. Michalides, H. Mineta et al., “Ampli- [39] A. F. L. Waitzberg, S. Nonogaki, I. N. Nishimoto et al., “Cli- nical significance of c-myc and p53 expression in head and fication of cyclin D1 in squamous cell carcinoma of the head and neck and the prognostic value of chromosomal abnormal- neck squamous cell carcinomas,” Cancer Detection and Preven- ities and cyclin D1 overexpression,” Cancer,vol. 79, no.2,pp. tion, vol. 28, no. 3, pp. 178–186, 2004. [40] M. Someya, K. I. Sakata, Y. Matsumoto et al., “The association 380–389, 1997. of DNA-dependent protein kinase activity with chromosomal [57] R. J. A. M. Michalides,N.M.J.Van Veelen,P.M.P.Kristel instability and risk of cancer,” Carcinogenesis,vol. 27, no.1,pp. et al., “Overexpression of cyclin D1 indicates a poor prognosis 117–122, 2006. in squamous cell carcinoma of the head and neck,” Archives [41] R. S. Bernard, L. Zheng, W. Liu, D. Winer, S. L. Asa, and S. of Otolaryngology-Head and Neck Surgery, vol. 123, no. 5, pp. Ezzat, “Fibroblast growth factor receptors as molecular targets 497–502, 1997. 8 Journal of Oncology [58] J. A. Akervall, Y. Jin, J. P. Wennerberg et al., “Chromosomal abnormalities involving 11q13 are associated with poor prog- nosis in patients with squamous cell carcinoma of the head and neck,” Cancer, vol. 76, no. 5, pp. 853–859, 1995. [59] J. G. Izzo, V. A. Papadimitrakopoulou, X. Q. Li et al., “Dys- regulated cyclin D1 expression early in head and neck tumori- genesis in vivo evidence for an association with subsequent gene amplification,” Oncogene, vol. 17, no. 18, pp. 2313–2322, [60] H. J. Roh, D. M. Shin, J. S. Lee et al., “Visualization of the tim- ing of gene amplification during multistep head and neck tu- morigenesis,” Cancer Research, vol. 60, no. 22, pp. 6496–6502, [61] D. L. Uhlman, G. Adams, D. Knapp, D. M. Aeppli, and G. Niehans, “Immunohistochemical staining for markers of fu- ture neoplastic progression in the larynx,” Cancer Research, vol. 56, no. 9, pp. 2199–2205, 1996. [62] V. A. Papadimitrakopoulou, J. Izzo, J. Keck et al., “Cyclin D1 and p16 alterations in advanced premalignant lesions of the upper aerodigestive tract: role in response to chemoprevention and cancer development,” Clinical Cancer Research, vol. 7, no. 10, pp. 3127–3134, 2001. [63] Y. Yamada, R. Pannell, A. Forster, and T. H. Rabbitts, “The oncogenic LIM-only transcription factor Lmo2 regulates an- giogenesis but not vasculogenesis in mice,” Proceedings of the National Academy of Sciences of the United States of America, vol. 97, no. 1, pp. 320–324, 2000. [64] N. Yamagata, K. Furuno, M. Sonoda, H. Sugimura, and K. Yamamoto, “Stomach cancer-derived del223V-226L mutation of the STCH gene causes loss of sensitization to TRAIL-med- iated apoptosis,” Biochemical and Biophysical Research Com- munications, vol. 376, no. 3, pp. 499–503, 2008. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal

Journal of OncologyHindawi Publishing Corporation

Published: Apr 12, 2012

References