TRENDS IN DIAGNOSTIC PATHOLOGY

New ideas regarding the use of diagnostic pathology in the evaluation of head and neck cancer parallel the introduction and use of novel techniques in other areas of diagnostic pathology. An increase in the use of molecular technology and the associated benefits that genetic information provides about the course of disease has had a definite effect on how pathologists and clinicians view disease. The opportunity to investigate the risk of adverse genetic events has greatly improved as knowledge of molecular biology grows. Traditional methods continue to be used to characterize oncologic processes. These morphologic approaches include electron microscopy, immunohistochemistry, and conventional histochemical staining. Although these methods provide diagnostic and prognostic information, they are being augmented with techniques that provide information on the genetic changes present within a tumor and the oncogenes and antioncogenes that influence biologic behavior. Use of these novel methods has contributed important information to the understanding of cellular differentiation and neoplastic development. As originally anticipated, not all information gained from these studies has been useful in defining the course of disease. Despite these limitations, valuable diagnostic and prognostic information has been gained with molecular technology. This information is currently being applied in protocol studies involving therapy for various malignant tumors, such as oncogene detection to define the course of squamous cell carcinoma of the head and neck. This chapter concerns trends in diagnostic pathology, provides current information on the molecular biologic aspects of head and neck cancer, and focuses on the use of molecular biologic investigations of this neoplastic process.
HUMAN GENOME PROJECT AND TECHNOLOGICAL DEVELOPMENT
The Human Genome Project was initiated in the 1990s to develop a comprehensive genetic and physical map of the human genome and to elucidate the complete DNA sequence of all human chromosomes. This project promises to provide new insight into the diagnosis and management of malignant diseases that affect humans. A highlight of the importance of this ambitious project is that it may become possible to obtain a genetic profile of all humans at birth. Issues related to the ethical use of this information have to be established, and use of this information has to be incorporated into standards of practice. In addition to identifying genes associated with cancer and other diseases, the Human Genome Project has led to the introduction of new terminology, contributed to the development of new technology, and provided novel ways to study cancer and other diseases. Genomics is the detection of genes associated with cancer and other diseases. Proteomics was introduced as a way to describe the study of the function of individual genes within the context of all genes in the cell at the protein level (functional genomics) (1). The proteome is defined as the expressed protein complement of a genome (2). The goal of proteomics is to develop a comprehensive, quantitative description of protein expression, which may include changes that occur during the development of tumors, dilated cardiomyopathy, or infectious disease and changes that occur after therapeutic intervention (2). Functional genomics also includes several terms such as transcriptome and physiome.
Laser capture microdissection was developed with the Human Genome Project (3). This was fortuitous because laser capture microdissection provides a mechanism whereby individual cells or groups of cells within tissues can be selectively removed and used for genetic analysis. The development of mouse models to study the effects of gene deletions and alterations or augmentation and the development of DNA microarray biotechnology concomitantly increased understanding and knowledge of mutated sequences in human tumors and their phenotypic expression. In concert with these methods or techniques, developments in computer technology and bioinformatics have contributed to the ability to evaluate the volume of data generated by these new technologies (4). The information that emerges from clinical trials will determine whether these studies improve understanding of the molecular anatomic and physiologic characteristics of normal and neoplastic cells. For example, investigators using a “lymphochip” to study lymphoma detected the complementary DNA (cDNA) arrays derived from mature lymphocytes and their precursors with the aim of determining the phenotypic expression of DNA alterations and the histologic type of lymphoma. With DNA microarray technology to analyze diffuse large cell lymphoma, two diverse phenotypes were found and were shown to have a profound influence on survival. Tumors with a profile of germinal B cells had a better overall response to treatment than tumors in which gene expression revealed activated B cells (5).
Information gained in the analysis of various types of tumors requires data from numerous patients. Individual differences, tumor heterogeneity, and novel methods to incorporate these findings into the current understanding of the multistage theory of cancer are necessary to ascribe genetic significance to the findings about a given tumor type. The goal of these studies is to tailor treatment to the genetic profile of a tumor. Paramount to these studies, however, is the need for continued advances in bioinformatics (4), the goal of which is to provide methods sufficient for data normalization and standards to provide statistical evaluation of the myriad data derived from these new technologies.
The use of laser capture microdissection in proteomics and genomic research has provided a method to selectively capture for analysis individual cells or groups of cells within tissues. In combination with DNA array technology (gene chip technology), in which large numbers of nucleic acid samples can be assayed, new advances are being made in the understanding of diseases that affect humans. With DNA chip technology, cDNA clone inserts are robotically printed onto a glass slide. They are subsequently hybridized to two different fluorescent labeled probes. The probes are pools of cDNA generated after isolation of messenger RNA (mRNA) from cells or tissues for comparative evaluation. The DNA probes are used to interrogate target sequences on the basis of specificity of hybridization to the known probe. The intensity and ratio of the fluorescent tag are measured, and the differences between the controls and the test samples are calculated to identify genes of importance in the test samples. An advantage of this technology is that it has produced a powerful method to evaluate the genetic composition of tissues from archival material and to document the genetic composition of tissues obtained from patients in clinical trials. The aim of this technology is to describe the multitude of genes expressed in a tumor, to develop genetic profiles of cancer among humans, and to tailor treatment to the genetic changes identified in a tumor sample.
p53 AND HEAD AND NECK CANCER
Molecular evaluation of malignant tumors that affect humans has included studies of cellular proliferation and oncogenesis in a variety of tumors, including tumors of the breast and prostate of adults and small blue round cell tumors of children. Most studies of oncogenes in head and neck cancer show a limited relation between oncogene activation and prognosis. No oncogene has achieved overall important measured against commonly used prognostic features. In some studies, however, when detection of an oncogene was combined with other prognostic indicators, a relation was shown between the presence of an oncogene and development and progression of cancer.
Considerable knowledge of the genetic nature of the biologic characteristics of tumors has emanated from studies of p53 (6). This tumor suppressor gene is located on the short arm of chromosome 17 (17p13.1). Wild-type p53 has a role in preventing accumulation of genomic abnormalities within cells that may lead to the development of a malignant phenotype. Mutations in p53 have been described in a variety of tumors, and this gene is commonly mutated in cancers that affect humans, including tumors of the head and neck region. Cytogenetic, molecular, and immunohistochemical methods have been used to study the role of p53 in carcinogenesis. Loss of the suppressor function of p53 most often is caused by complete loss of one allele associated with a point mutation in the second allele. The resultant mutated gene lacks the suppressor activity of the wild-type gene, is metabolically stable, and has a long half-life. Under normal conditions, p53 has a short half-life and may not be detected with current immunohistochemical methods. When p53 is mutated, altered forms can be found in 30% to 80% of tumors. The altered forms are more stable and are therefore easy to detect with immunohistochemical methods. p53 normally prevents cells with damaged DNA from progressing through the cell cycle in the transition from G1 into the S phase. This process allows the cell time to repair DNA damage. The importance of a functionally intact G1 cell cycle checkpoint is emphasized by the fact that cells lacking in wild-type p53 protein enter the S phase without having repaired the DNA. The result is progressive genomic instability followed by initiation of the malignant process. p53 acts as a transcription promoter and interacts with cellular proteins such as CCAAT-binding protein and the protein product mdm2 (7). Tumors associated with an abnormal p53 gene have been reported to be of high histologic grade and to have increased proliferative activity. In some studies, p53 mutations have been associated with shorter disease-free intervals and poor overall survival.
The role of p53 in the development of squamous cell carcinoma of the head and neck is not well established (Table 7.1). Reports in the literature support a role of this tumor suppressor gene in the evolution of cancer (8). However, in neoplasms involving the head and neck, the relation is unclear. Hamel et al. (9) reported that patients homozygous for the arginine allele at codon 72 of p53 had an increased risk of cervical cancer related to infection with the human papillomavirus (HPV). Despite the recognized association between epithelial cancer of the uterine cervix and HPV infection, no association was found between HPV infection and squamous cell carcinoma of the head and neck in an analysis of 163 cases. Other studies of oral squamous cell carcinoma and squamous cell carcinoma of the head and neck had similar findings (10). Of interest is the potential relation of cyclin D1 to the development of multiple primary neoplasms not associated with p53, the decreased median relapse-free survival time for p53-negative tumors, and the absence of a positive correlation between the Bcl-2 family of proteins and p53 in the genesis of tumors of the head and neck.

Gleich et al. (10) suggested that the genesis of squamous cell carcinoma of the head and neck is most likely mediated by a variety of pathways and that single genetic alterations are not sufficient to influence survival. Patients with tumors that have one genetic loss had a 2-year survival rate of 78%. Patients with tumors that had two or more genetic alterations had a median survival rate of 58%. Mutation of the p53 gene was not associated with survival but was believed to represent a clonal marker not susceptible to change during metastasis (11). Warnakulasuriya (12) stated that p53 mutations are not useful in predicting outcome for patients with oral leukoplakia and are not informative as a sole marker to predict tumor development among persons at high risk. Some studies, however, have shown that p53 has prognostic utility in predicting the biologic behavior of squamous cell carcinoma of the tongue. Unal et al. (13) suggested that p53 may have a role in the biologic behavior of squamous cell carcinoma of the tongue. They reported that p53 immunoreactivity correlates with tumor size, lymph node metastasis, and stage.
Kudo et al. (14) investigated the possible association between p53 and p21 (cyclin-dependent kinase inhibitor) in the development of oral epithelial dysplasia and squamous cell carcinoma. No association was found, but the authors suggested that the combination of p21 and p53 expression may play a role in prognosis among patients with oral dysplasia and carcinoma. Expression of p21 in oral squamous cell carcinoma may be related to cellular proliferation and mdm2 expression that is independent of p53 protein alterations. Lam et al. (15) found that p21 is associated with tumor stage, tumor grade, nodal status, and mitotic count. Their findings showed that p21 is an important factor in the progression of squamous cell carcinoma of the larynx and esophagus. Expression of p53, p21, Rb, and mdm2 proteins in carcinoma of the tongue was investigated in a study involving patients younger than 35 years and patients older than 75 years. The results suggested that there are no differences in expression of these gene products in carcinoma of the lateral aspect of the tongue (16). In laryngeal carcinoma, p27 expression was found to be an independent prognostic indicator. In predicting the development of cancer among patients with oral leukoplakia, several factors have been found to be associated with the development of cancer. These include oral histologic findings, cancer history, chromosomal polysomy, p53 protein expression, and loss of heterozygosity at chromosomes 3p and 9p.
MICROSATELLITE INSTABILITY AND HEAD AND NECK CANCER
Little information exists about the relation between microsatellite instability and oropharyngeal carcinoma. Lynch and Kaul (17) discussed microsatellite instability in colorectal carcinoma in an editorial accompanying a published report that documented the influence of microsatellite instability on the development of colorectal carcinoma. It was suggested that the colorectal carcinomas that had microsatellite instability were more likely to be indolent. However, the results were not statistically significant when compared with the Dukes classification of colorectal cancer. One study (1 showed that chromosome tetraploidization is important in malignant transformation of laryngeal tumors. In this study, most dysplastic lesions and carcinomas in situ contained chromosomal abnormalities. The time to development of cancer from baseline biopsy was shorter among patients with unstable chromosomal contents than among the group with stable chromosome contents.
Evaluation of 51 squamous cell carcinomas from various sites in the head and neck area revealed that overexpression of p53 correlated with an increased prevalence of chromosomal abnormalities and aneuploid tumor. A significant correlation was shown between tumors that had metastasized and ploidy. These findings showed that tumors with high rates of metastasis had increased chromosomal imbalances. Some studies have shown a poor correlation between microsatellite instability and risk factors associated with squamous cell carcinoma of the head and neck. On chromosome 9p, loss of heterozygosity targets the same region as documented in other tumor types. The suggestion is that the patterns of microsatellite instability documented in other types of tumors are similar. To document the role of loss of heterozygosity, 77 oral squamous cell carcinoma with 11 microsatellite markers located on chromosomes 3p and 9p were studied (26). Loss of heterozygosity was identified in multiple sites, and 44% of the tumors showed allelic loss at one or more loci on both 3p and 9p. No correlation was shown between the frequency of loss of heterozygosity and stage of disease.
HUMAN PAPILLOMAVIRUS AND SQUAMOUS CELL CARCINOMA OF THE HEAD AND NECK
Human papillomavirus has a role in the genesis of squamous cell carcinoma of the head and neck (19,20). Approximately 20% of oropharyngeal tumors had the same type of or DNA similar to the type present in squamous cell carcinoma of the uterine cervix, perianal and anal skin, vulva, and penis. The results of these studies suggested that HPV might have a causal relation in some types of head and neck cancer. These results also suggested that HPV-positive tumors arising in the head and neck area have an improved prognosis (19).