Mouth cancer for clinicians part 1: cancer

What is cancer?

The word tumour in Latin means a swelling but a tumour or swelling is not always a cancer. Some tumours may be caused by inflammation, infections, cysts or fluid-filled lesions or be due to new growths (neoplasms). There are several causes of non-cancerous (benign) neoplasms in the mouth and oropharynx. Benign neoplasms do not invade other organs or spread to other parts of the body, although they can grow to a large size and start to press on surrounding organs and tissues and cause ill-effects. In contrast, those neoplasms that are capable of invading and spreading are referred to as malignant tumours or malignant neoplasms or ‘cancer’ – they are autonomous and have the capacity to grow rapidly and to metastasize or spread to other tissues.

Cancer, however, is a term that refers to a group of more than a hundred diseases that can originate in many different parts of the body. Most malignant neoplasms grow as solid masses of tissue; some, like leukaemias, grow as cell suspensions but the following features are common to all types of cancer:

If cancer is left untreated, it can invade tissues, spread through the body and eventually lead to death.

How does cancer arise?

The trillions of cells that grow, divide and die normally do so in a very organized manner. These processes are tightly controlled by DNA (deoxyribonucleic acid) – the ‘blueprint’ that instructs cell growth, protein synthesis, function and death. If DNA becomes damaged in a normal cell, that cell then either repairs the damage or the cell dies. DNA damage is often termed a ‘mutation’.

DNA damage is often a mutation which leads to altered amino acids and thus changes in cell proteins and activities that may lead to the complex pathophysiology of cancer. Gene changes may be a single point mutation affecting a single DNA nucleotide, or can occur at many levels (eg a gain or loss of entire chromosomes or deletions and insertions, especially at the promoter region of the gene). Disruption of a single gene may also result from integration of genomic material from a DNA virus (eg certain human papillomaviruses, HPV) or retroviruses, which may lead to the formation of oncogenes (cancer-promoting genes). Translocation is yet another process when two separate chromosomal regions become abnormally fused, often at a characteristic location. One example is the Philadelphia chromosome, a translocation of chromosomes 9 and 22, seen in chronic myeloid leukaemia, which results in the production of an oncogenic enzyme called tyrosine kinase (BCR-abl fusion protein). Cancer cell growth differs from normal cell growth. Instead of dying, the cancer cells continue to grow and form new, abnormal cells, which can also invade other tissues.

In order to transform a normal cell into a cancer cell, changes in many genes are typically required (as discussed below), but there are two broad categories of genes which are especially involved:

The genetic abnormalities in mouth cancer are many, and may alter the protein coded for by particular genes and modify cell behaviour. They are mainly at chromosomes 9p21 (TSG is p16), 3p, 17p and 8p, and include mutation of the TSGs p53, retinoblastoma (Rb) and p16 and amplification of cyclin D1. Most mouth cancers have copy number alterations (CNAs), including losses of chromosomal regions 3p and 8p, and gains of 3q, 5p and 8q.

What is responsible for the DNA damage?

DNA mutation is most commonly spontaneous and it is now suggested that the majority are due to ‘bad luck’, that is, random mutations arising during DNA replication in normal, non-cancerous stem cells. DNA damage may, on rare occasions, be inherited from parents, such as in some breast cancers, or in the head and neck cancers seen in Fanconi anaemia. However, mutations may also be triggered by exposure to certain exogenous factors (mutagens or carcinogens) – termed ‘risk factors’ – such as those present in tobacco and alcoholic beverages. Malignant transformation may follow an accumulation of exposure to such mutagens and, along with damage from spontaneous mutations, the subsequent accumulation of genetic changes or molecular aberrations. Working in the other direction are factors that may confer some protection (Figure 2).

Mutations may affect cell behaviour:

Several DNA mutations are necessary before the affected cells change in appearance and behaviour to a recognizably pre- or potentially malignant cell characterized by an ability to proliferate in a less-controlled fashion than normal (they become autonomous). The progression from a normal cell to a pre-malignant or a potentially malignant cell – oncogenesis (carcinogenesis) – is characterized by an ability of cells to escape normal growth control mechanisms, and to proliferate autonomously. A series of steps lead to the aberrant expression and function of molecules regulating cell signalling, growth, survival, motility, angiogenesis (blood vessel proliferation), and cell cycle control. Crucial genes involved in cell cycle regulation are those encoding proteins in the p53 and Rb pathways (see below). Normally, the body safeguards against cancer in a number of ways, such as: apoptosis (programmmed cell death) or a process by which abnormal cells die of their own accord, helper molecules (some DNA polymerases), possibly senescence or ageing, etc. A key cellular function that is often changed in cancer cells to overcome senescence and to obtain limitless replicative potential is the regulation of the cell cycle. In cancer cells, the damaged DNA is not repaired, and neither does the cell die, but rather it gives rise to more such abnormal cells with abnormal DNA. These new cells all have the same defective DNA of the original cancer cell.

The effects of these changes may eventually be visible under the microscope as dysplasia – disordered cell size and arrangement, with abnormal cell divisions (mitoses). Dysplasia can transform to cancer, characterized by malignant epithelial cells (keratinocytes) which proliferate and invade across the epithelial basement membranes as a ‘growth’ into the underlying tissues. The changes may become evident clinically, usually as a red (erythroplasia) or white lesion (leukoplakia).

Clinical assessment of cancer risk in oral potentially malignant disorders (PMD) (Article 6) is most inaccurate and, although the World Health Organization has criteria to define dysplasia, it may be difficult to make a confident objective categorization of dysplasia owing to a high inter-observer and intra-observer variation in dysplasia assessment. Molecular aberrations have been identified in PMD which increase in number as lesions progress towards malignancy. These cancer-associated genetic changes in PMD are increasingly being used to discriminate leukoplakias with a low risk from those with a high risk of malignant transformation.

Development of technologies like immunohistochemistry, flow cytometry and molecular biologic approaches to cancer diagnosis have profoundly influenced knowledge. Microarray DNA technology has shown that many genes can be involved in oncogenesis. RNA and DNA profiling studies, in particular, have highlighted the molecular heterogeneity of these lesions.

What are the main mouth cancer histological and molecular changes?

Several DNA mutations are necessary before the affected cells change appearance and behaviour to a recognizably pre- or potentially malignant cell characterized by an ability to proliferate in a less-controlled fashion than normal (they become autonomous). The effects of these changes may be seen under the microscope as dysplasia. This can transform to cancer, characterized by malignant epithelial cells (keratinocytes) which proliferate and invade across the epithelial basement membranes as a ‘growth’ into the underlying tissues (Figure 3).

Solid malignant neoplasms are composed of a parenchyma that contains cancer tissues and cells, and the other is a stroma that the neoplastic cells induce and in which they are dispersed. All solid tumours require stroma if they are to grow beyond a minimal size; the stroma is essential for malignant neoplasm growth and contains non-malignant supporting tissue such as connective tissue, blood vessels and, very often, inflammatory cells. In addition, cancers also have the property of new blood vessel formation (neoangiogenesis). There are in addition proteoglycans and glycosaminoglycans, interstitial collagens (types I, III and, to a lesser extent, type V), fibrin, fibronectin, fibroblasts, etc. Malignant neoplasms that originate from epithelial cells (carcinomas) have a basal lamina that separates malignant cells from stroma. However, the basal lamina is often incomplete, especially at points of invasion. Indeed, invasion of the epithelial basement membrane is the feature that distinguishes a PMD from cancer. The number of lymph node metastases, the proportion in the neck and extra-nodal spread are important prognostic factors and predictors of distant disease and survival. Metastatic dissemination involves several steps, including degradation of the extracellular matrix as one of the initial steps. Matrix metalloproteinases (MMPs) are known to be involved in the degradation of the extracellular matrix. The CSMD1 gene on chromosome 8p or other genes or micro-RNAs (miRNAs) might be involved in invasion and metastasis. All solid tumours induce neo-angiogenesis, usually by producing angiogenic factors (growth factors that induce sprouting of endothelial cells, and formation of new vessels feeding the tumour). The strongest angiogenesis inducer is Vascular Endothelial Growth Factor (VEGF), the expression of which is a poor prognostic factor.

Loss of heterozygosity (LOH) at chromosomes 3p, 9p and 17p occur in dysplasia, apparently reflecting early carcinogenesis, whereas other alterations at chromosomes 11q, 4q and of chromosome 8 are typically present in carcinomas, probably corresponding to a relatively late phase in carcinogenesis. LOH at chromosome 9p21 is thought to be an early event in mouth carcinogenesis. Cyclin dependent kinase inhibitor 2A (also known as p16Ink4A or p16 [CDKN2A]), which encodes the TSG p16INK4A, is located on chromosome 9p21 and is frequently inactivated in oral squamous cell carcinoma (OSCC). The loss of chromosome 9p21 (which is where CDKN2A is located) and p53 mutations are frequently found in PMD. Mutations in TSGs p53 and Rb are common in tobacco-related malignant oral lesions. CCND1, which encodes cyclin D1, on chromosome 11q13, is amplified or gained in most HPV-negative OSCC. Together with the abrogation of p53, these changes cause cellular immortalization (Figure 4). Tumours with p53 mutation-positive molecular fields have a 10% increase in local recurrence rate compared to those with wild type (non-mutated) p53.

The normal shortening of telomeres (ends of chromosomes) also needs to be overcome to achieve the limitless cell replicative potential in cancers. Abnormal expression of the enzyme telomerase reverse transcriptase (TERT), which controls telomeres, occurs, as well as somatic mutations in p53, overexpression of cyclin D1, or a p16INK4A-insensitive cyclin-dependent kinase 4 (CDK4) mutant. Telomerase or TERT activity is seen in most OSCCs.

Changes in growth factor signalling are also important. One growth factor, the epidermal growth factor receptor (EGFR) contains receptor tyrosine kinases of the Erb-b family which initiate a signalling cascade (through other pathways [Ras–MAPK, PI3K–PTEN–AKT and phospholipase C]). EGFR is also able to translocate to the nucleus and induce CCND1, has been implicated in the malignant transformation of oral keratinocytes, and is often overexpressed in OSCCs. These findings have led to EGFR-specific antibodies (cetuximab) now being used for targeted mouth cancer therapy (Article 12).

Other factors affecting cell signalling include the transforming growth factor-β (TGFβ) pathway – an inhibitory growth factor – and the PI3K–PTEN–AKT pathway, which evades apoptosis and mutations may cause cancer cell increased migration and invasion.

Smoking-related (HPV-unrelated) HNSCCs typically have loss-of-function p53 mutations and CDKN2A inactivation, with amplification of 3q26/28 and 11q13/22. Indeed, TP53 mutations are detected in all HPV-negative mouth cancer cases and there is abrogation of the cell cycle G1/S checkpoint via CDKN2A/B deletion and/or CCND1 amplification in most HPV-tumours, suggesting that, in future, CDK inhibitors may have possible therapeutic uses.

In contrast, most HPV-related oropharyngeal cancers have wild-type TSGs Rb and p53 that are also the first to be inactivated by the viral E6 and E7 oncogenes found in oncogenic HPVs (Article 5). HPV-associated cancers are characterized by amplification of the cell cycle gene E2F1, mutations of PI3 kinase (PIK3-CA), and loss of TRAF3. Mutation and copy number alterations of PI3K pathway components appear particularly common, further supporting a role for HPV inhibiting the p53 and RB pathways. HPV also modulates the epigenome via hypermethylation of genes of the Polycomb repressive complex 2 (eg cadherins), implicated in cancer progression and metastasis.

Recurrences and second primary tumours

There is a high propensity to develop local recurrences after treatment of oral squamous cell carcinoma (OSCC) and a high likelihood that multiple independent tumours will develop in the head and neck mucosa. It is also known that, in at least 35% of OSCCs, the oral carcinoma is surrounded by mucosal epithelium that contains genetic changes which, though macroscopically normal, may be histologically dysplastic. This tumour-adjacent mucosal epithelium characterized by genetic changes has also been termed ‘field’ changes and, if left after resection, may readily initiate recurrence.

Furthermore, people with one particular cancer are also predisposed to develop another malignant neoplasm – a second primary cancer (second primary tumour; SPT). In the case of mouth cancer, SPTs may be in the mouth, or aerodigestive tract (pharynx, larynx, bronchi, oesophagus). This has been termed ‘field cancerization’ and is linked to the frequent observation of dysplastic changes surrounding these tumours with the occurrence of local recurrences and multiple primary tumours. This concept has now been defined in molecular terms on the basis of the genetic characterization of morphological changes in the squamous epithelium. SPTs are usually in the upper aerodigestive tract: women have a higher risk of a SPT than do men.

Summary

Head and neck squamous cell cancers (HNSCC), the most common of which are mouth cancers, arise as a consequence of DNA mutations caused mainly by free radicals and oxidants, and DNA mutations change various crucial genes and other nucleic acid components involved in cell growth and control (and are potential targets for anti-cancer therapies) (Figure 5) such as:

Oncogenesis (carcinogenesis) is the progression from a normal cell (eg epithelial cell or keratinocyte) to a pre-malignant or a potentially malignant cell, characterized by a series of steps leading to the aberrant expression and function of molecules regulating cell signalling, cell cycle control, growth, survival, motility and angiogenesis (new blood vessel proliferation).

Cell cycle control is disturbed particularly by over-expression or over-activity of oncogenes, such as the EGFR gene. Cell cycle control is also influenced by TSGs, which help protect cells but, if a TSG is defective, cancer protection is impaired.

DNA technology is making major advances in understanding the pathogenesis and suggesting routes towards developing targeted treatments. Microarray DNA technology has shown that many genes can be involved in oncogenesis and that some genes (eg EGFR) may be potential targets for cancer therapy (‘targeted therapy’; Article 12). Single nucleotide polymorphisms (SNPs) – gene areas with altered DNA sequences which may not lead to an amino acid alteration – may, in various genes (TSGs, xenometabolizing enzymes, and DNA repair), also sometimes play a role in cancer. Information from the human genome project together with technological advances has allowed the development of rapid, accurate platforms to carry out comprehensive genetic characterization of germ line and tumour samples, technology referred to as next-generation sequencing (NGS).

What are head and neck cancers?

Cancers can arise in any tissues or organs in the head and neck: there are over 30 different sites that can be affected. Most are squamous cell carcinomas. In the head and neck region the mouth is the most common site. Head and neck cancers can also affect the throat, and there are rarer cancers of the nose, sinuses, salivary glands, skin and middle ear, or arising in other tissues (Figure 6).

What other malignant tumours affect the mouth or oropharynx?

One in 10 mouth and oropharyngeal cancers (10%) are one of the following types:

What are mouth cancers?

Cancers can develop on the lips, tongue, floor of the mouth, the buccal mucosae, the palate, the gingivae or oropharynx. Mouth cancers (Figure 7) are mostly squamous carcinomas and include:

There are also malignant tumours that arise from other structures including the salivary glands and paranasal air sinuses, or other tissues, but these cancers do not share a similar natural history and are not included here.

What are the types of mouth cancer?

Cancers can be described according to the cell type in which the cancer began. In the oral mucosa the most common arise from keratinocytes and are squamous cell carcinomas (SCC). About 9 out of 10 mouth cancers are squamous and are termed OSCC. About 1 in 20 mouth cancers (5%) are an unusual type of squamous cell carcinoma which rarely spreads to other parts of the body but can grow deeply into local tissues.

OSCC cancers are more common:

Oral pre-cancerous (potentially malignant) disorders include especially (Article 6):

What is the international classification of mouth tumours?

Of the many malignant neoplasms that can affect the mouth, oral squamous cell carcinoma (OSCC) is the most important.

Cancers of the ‘oral cavity and oropharynx’, as classified in the WHO International Classification of Diseases (ICD), include cancers of the lip, tongue and mouth (oral cavity) [ICD-10: C00-06], and oropharynx [ICD-10: C09-C10], but excludes the salivary glands [C07-08] and other pharyngeal sites [C11-13] (Table 1). ICD-9 is shown for comparison.

What is oropharyngeal cancer (cancer of the oropharynx)?

Oropharyngeal cancers are most commonly SCCs. Oropharyngeal cancer is uncommon: about 1,340 people are diagnosed each year in the UK. Like oral cancer, most oropharyngeal cancer affects mainly people over 50 and is more common in men than in women. However, oropharyngeal cancers are becoming more prevalent, which may be related to an increase in oropharyngeal viral infections (Article 4), notably HPV.

Conclusion

This review gives the dental team a simplified overview of carcinogenesis, and a summary of cancers that affect the oral region, as this is the single most important aspect of oral healthcare.