Benign and Malignant Tumors of the Oral Cavity

Clinical Features

The papilloma is an exophytic growth made up of numerous, small finger like projections which result in a lesion with a roughened, verrucous or ‘cauliflower like’ surface. It is nearly always a well circumscribed pedunculated tumor, occasionally sessile. It is painless, usually white but sometimes pink in color. Intraorally it is found most commonly on the tongue, lips, buccal mucosa, gingiva and palate, particularly that area adjacent to the uvula (Fig. 2-1). The majority of papillomas are only a few millimeters in diameter, but lesions may be encountered which measure several centimeters. These growths occur at any age and are seen even in young children. A series of 110 cases has been reported by Greer and Goldman, while a series of 464 cases has been studied by Abbey and his coworkers; this is somewhat indicative of the prevalent nature of the lesion.

The common wart or verruca vulgaris, is a frequent tumor of the skin analogous to the oral papilloma. It is uncommon on oral mucous membranes but extremely common on the skin. The associated viruses in verruca are the subtypes HPV-2, HPV-4 and HPV-40. Clinically it shows pointed or verruciform surface projections, a very narrow stalk, appears white due to considerable surface keratin, and presents as multiple or clustered individual lesions. It enlarges rapidly to its maximum size, seldom achieving more than 5 mm in greatest diameter. Verruca vulgaris is contagious and capable of spreading to other parts of an affected person’s skin or membranes by way of autoinoculation. Lesions that are histologically identical to the verruca vulgaris of the skin are frequently found on the lips and occasionally intraorally (Fig. 2-2). These are often seen in patients with verrucae on the hands or fingers, and the oral lesions appear to arise through autoinoculation by finger sucking or fingernail biting.

Papilloma like or papillomatous lesions as well as ‘pebbly’ lesions and fibromas of various sites in the oral cavity are recognized as one of the many manifestations of the multiple hamartoma and neoplasia syndrome (Cowden’s syndrome). This is an autosomal dominant disease characterized by facial trichilemmomas associated with the gastrointestinal tract, the thyroid, CNS and musculoskeletal abnormalities as well as oral lesions. It is also considered a cutaneous marker of breast cancer.

Histologic Features

The microscopic appearance of the papilloma is characteristic and consists of many long, thin, finger-like projections extending above the surface of the mucosa, each made up of a continuous layer of stratified squamous epithelium and containing a thin, central connective tissue core which supports the nutrient blood vessels (Fig. 2-1 C). Some papillomas exhibit hyperkeratosis, although this finding is probably secondary to the location of the lesion and the amount of trauma or frictional irritation to which it has been subjected. The essential feature is a proliferation of the spinous cells in a papillary pattern; the connective tissue present is supportive stroma only and is not considered a part of the neoplastic element. Occasional papillomas demonstrate pronounced basilar hyperplasia and mild mitotic activity which should not be mistaken for mild epithelial dysplasia. Koilocytes (HPV altered epithelial cells with perinuclear clear spaces and nuclear pyknosis) may or may not be found in the superficial layers of the epithelium. The presence of chronic inflammatory cells may be variably noted in the connective tissue.

Treatment and Prognosis

Treatment of the papilloma consists of excision, including the base of the mucosa into which the pedicle or stalk inserts. Removal should never be accomplished by an incision through the pedicle. If the tumor is properly excised, recurrence is rare. The possibility of malignant degeneration in the oral papilloma is quite unlikely, although fixation of the base or induration of the deeper tissues should always be viewed with some suspicion.

Intraoral verruca vulgaris is also treated effectively by conservative surgical excision or curettage, but liquid nitrogen cryotherapy and topical application of keratinolytic agents (usually containing salicylic acid and lactic acid) are also effective. Recurrence is seen in a small proportion of treated cases.

Clinical Features

Keratoacanthoma has been reported in all age groups, but incidence increases with age. It occurs twice as frequently in men as in women and is less common in darker skinned individuals. Most keratoacanthomas occur on sun-exposed areas. The face, neck, and dorsum of the upper extremities are common sites; 8.1% of the cases occurred on the lips and the vermilion border of both the upper and lower lip are affected with equal frequency. Intraoral lesions are quite uncommon, a total of four cases were reviewed by Freedman and his associates.

Lesions typically are solitary and begin as firm, round, skin-colored or reddish papules that rapidly progress to dome-shaped nodules with a smooth shiny surface and a central crateriform ulceration or keratin plug that may project like a horn. The lesion appears as an elevated umbilicated or crateriform one with a depressed central core or plug (Fig. 2-3). It is seldom over 1.0 to 1.5 cm in diameter. The lesion is often painful and regional lymphadenopathy may be present.

The clinical course of the lesion is one of its unusual aspects. It begins as a small, firm nodule that develops to full size over a period of four to eight weeks, persists as a static lesion for another four to eight weeks, then undergoes spontaneous regression over the next six-to eight-week period by expulsion of the keratin core with resorption of the mass. However, lesions with an overall duration of as long as two years have been reported. Recurrence is rare.

The differential diagnoses include actinic keratosis, molluscum contagiosum, Muir-Torre syndrome, squamous cell carcinoma, and verrucous carcinoma.

Histologic Findings

The lesion consists of hyperplastic squamous epithelium growing into the underlying connective tissue. The surface is covered by a thickened layer of parakeratin or orthokeratin with central plugging. The epithelial cells do not usually show atypia but occasionally dysplastic features are found. At the deep leading margin of the tumor, islands of epithelium often appear to be invading and frequently this area cannot be differentiated from epidermoid carcinoma. Lapins and Helwig, among others, have also reported that the keratoacanthoma on occasion may invade perineural spaces, but this does not adversely affect the biologic behavior of the lesion; neither can it be used as a distinguishing characteristic between keratoacanthoma and epidermoid carcinoma. Pseud-ocarcinomatous infiltration typically presents a smooth, regular, well-demarcated front that does not extend beyond the level of the sweat glands. The connective tissue in the area shows chronic inflammatory cell infiltration. The most characteristic feature of the lesion is found at the margins where the normal adjacent epithelium is elevated towards the central portion of the crater; then an abrupt change in the normal epithelium occurs as the hyperplastic acanthotic epithelium is reached. For this reason, the diagnosis may be impossible if the adjacent border of the specimen is not included in the biopsy.

The term squamous cell carcinoma, keratoacanthoma type has been introduced for otherwise classic keratoacanthomas that reveal a peripheral zone formed by squamous cells with atypical mitotic figures, hyperchromatic nuclei, and loss of polarity to some degree. These marginal cells also may penetrate into surrounding tissue in a more aggressive pattern.

Treatment

The lesion is usually treated by surgical excision. Spontaneous regression does not occur in every case; where it has occurred, residual scar may be present. Prognosis is excellent following excisional surgery. Recurrent tumors may require more aggressive therapy. Patients with a history of keratoacanthoma should be followed for the development of new primary skin cancers (squamous cell carcinoma, in particular).

Clinical Features

Ainsworth and her colleagues separate the congenital nevi of the skin into ‘small’ and ‘garment’ nevi. The ‘small’ nevi are greater than 1 cm in diameter and usually 3–5 cm. The ‘garment’ nevi are greater than 10 cm in diameter and can cover large areas of the skin. The congenital nevi occur in 1–2.5% of neonates and, with the passage of time, may change from flat, pale tan macules to elevated, verrucous, hairy lesions. Approximately 15% occur on the skin of the head and neck. Intraoral occurrence is extremely rare.

Acquired nevi are extremely common. They appear in about the eighth month of life and increase in number with age, apparently reaching their peak numerically in the late third decade of life. Clark has stated that the number of nevi which a person has is genetically determined. Interestingly, the number of nevi begin to decrease as one ages, so that elderly persons average far fewer nevi than younger adults.

The intradermal nevus (common mole) is one of the most common lesions of the skin, most persons exhibiting several, often dozens, scattered over the body. The common mole may be a smooth flat lesion or may be elevated above the surface; it may or may not exhibit brown pigmentation, and it often shows strands of hair growing from its surface (Fig. 2-4). This form of mole seldom occurs on the soles of the feet, the palms of the hands or the genitalia.

The junctional nevus may appear clinically similar to the intradermal nevus, the distinction being chiefly histologic. It is extremely important; however, that a distinction be drawn, since the prognosis of the two lesions is different, as will be pointed out later.

The compound nevus is a lesion composed of two elements: an intradermal nevus and an overlying junctional nevus.

The spindle cell and/or epithelioid cell nevus (Spitz nevus) occurs chiefly in children, only about 15% appearing in adults, and may appear histologically similar to malignant melanoma in the adult. Seldom; however, does this lesion exhibit clinically malignant features. Essentially, this lesion is clinically benign, but histologically malignant. Weedon and Little have reviewed 211 cases of spindle cell and epithelioid cell nevi while Allen has reviewed 262 cases, and none of these authors have found any nevus of this type on a mucous membrane surface, including the oral cavity.

The blue nevus is a true mesodermal structure composed of dermal melanocytes which only rarely undergo malignant transformation. It occurs chiefly on the buttocks, on the dorsum of the feet and hands, on the face and occasionally on other areas. The majority of blue nevi are present at birth or appear in early childhood and persist unchanged throughout life. The lesion is smooth, exhibits hairs growing from its surface and varies in color from brown to blue or bluish black. Scofield reviewed this lesion in 1959 and reported two intraoral cases, the first recorded occurrence of the lesions in the mouth. Since then, numerous intraoral lesions have been reported.

Histologic variants of acquired nevi such as neural, balloon cell and halo nevi occur. Only the halo nevus is clinicopathologically distinctive. The neural and balloon cell nevi are histologic variants of intradermal or compound nevi.

Oral Manifestations

Eight-five percent of oral nevi are found in patients younger than 40 years. Oral nevi are found in all races and more frequently in whites, in whom 55% of reported oral nevi occurred. Approximately 23% of oral nevi occurred in black patients. Patients with junctional or compound nevi were relatively young; they were aged 22 and 24 years, respectively, whereas the mean age of patients with intramucosal nevi and blue nevi was 35 years and 38 years, respectively. Studies have revealed that oral mucosal nevi are slightly predominant in women rather than men. Oral nevi most commonly occur on the hard palate, with almost 40% presenting in that location. The second most common location is the buccal mucosa (20% of cases). Other common locations include the vermilion border of the lip and the labial mucosa. Roughly 10% of all types of oral nevi are found on the gingiva. Only one mucosal nevus has been reported on the tongue or floor of the mouth.

Most oral nevi are asymptomatic, and the lesions are usually detected as an incidental finding on routine dental examination. Nevi are often mistaken for melanotic macules, amalgam tattoos, physiologic ethnic pigmentation, smoker’s melanosis, or other vascular or pigmented lesions.

An aid in differential diagnosis is that melanotic macules and amalgam tattoos are usually flat while the majority of nevi elevate from the mucosal surface. Ethnic pigmentation is nearly always symmetric and rarely affects the surface topography or disturbs the normal stippling in the gingiva. Smoker’s melanosis involves only the anterior gingiva and most often occurs in women who smoke and take oral contraceptives. Vascular lesions can be mistaken for melanocytic proliferations; the former usually blanch with compression, and aspiration may be helpful in differentiating a nevus from a vascular process. Malignant melanoma is frequently associated with diffuse areas of pigmentation, possible ulceration, nodularity, variegation in color, and an irregular outline.

Approximately 85% of oral nevi are pigmented. Pigmentation usually varies from brown to black or blue. Nevi are well circumscribed, round, or oval, and are raised or slightly raised in 65–80% of cases. Approximately 15% of nevi are amelanotic. Many amelanocytic lesions appear as sessile growths that resemble fibromas or papillomas or excrescences of normal color.

The anatomic distribution of nevi closely follows its histologic type. Almost two-thirds of blue nevi occur in hard palate. Intramucosal nevi are distributed almost equally between the hard palate and buccal mucosa, with almost 25% in each location. Approximately 17% of intramucosal nevi are on the gingiva, 12% on the vermilion border of the lip, and almost 9% on the labial mucosa.

Histologic Features

The nevus cells are assumed to be derived from neural crest and their envisaged relationship with true melanocytes remains uncertain. Nevus cells are large ovoid, rounded, or spindle-shaped cells with pale cytoplasm; and may contain granules of melanin pigment in their cytoplasm. The nucleus is vesicular and lacks the dendritic processes typical of melanocytes. In addition, they either lack contact inhibition or lose it shortly after the proliferation process begins. Melanosomes are retained by nevus cells and are not transferred to adjacent keratinocytes. They tend to be grouped in sheets or cords which are called nest or thèque. Nevus cells also have the ability to migrate from the basal cell layer into the underlying connective tissue.

Multinucleated giant nevus cells are sometimes seen, but are of little prognostic significance. Mitotic figures are not common.

In the intradermal nevus, the nevus cells are situated within the connective tissue and are separated from the overlying epithelium by a well defined band of connective tissue. Thus, in the intradermal nevus, the nevus cells are not in contact with the surface epithelium (Fig. 2-5 A).

In the junctional nevus, this zone of demarcation is absent, and the nevus cells contact and seem to blend into the surface epithelium. This overlying epithelium is usually thin and irregular and shows cells apparently crossing the junction and growing down into the connective tissue—the so-called abtropfung or ‘dropping off ’ effect. This ‘junctional activity’ has serious implications because junctional nevi have been known to undergo transformation into malignant melanomas (Fig. 2-5 D).

The compound nevus shows features of both the junctional and intradermal nevus. Nests of nevus cells are dropping off from the epidermis, while large nests of nevus cells are also present in the dermis (Fig. 2-5 B, C).

The spindle cell and/or epithelioid cell nevus is commonly composed of pleomorphic cells of three basic types: spindle cells, oval or ‘epithelioid’ cells, and both mononuclear and multinucleated giant cells. These are arranged in well-circumscribed sheets and there is generally considerable junctional activity.

The blue nevus is of two types: the common blue nevus and the cellular blue nevus. In the common blue nevus, elongated melanocytes with long branching dendritic processes lie in bundles, usually oriented parallel to the epidermis, in the middle and lower third of the dermis. There is no junctional activity. The melanocytes are typically packed with melanin granules, sometimes obscuring the nucleus, and these granules may extend into the dendritic processes. In the cellular blue nevus, an additional cell type is present: a large, round or spindle cell with a pale vacuolated cytoplasm. These cells commonly are arranged in an alveolar pattern.

Treatment and Prognosis

Since the acquired pigmented nevus is of such common occurrence, it would obviously be impossible to attempt to eradicate all such lesions. It has be-come customary to recommend removal of pigmented moles if they occur in areas which are irritated by clothing, such as at the belt or collar line, or if they suddenly begin to increase in size, deepen in color, or become ulcerated. Allen and Spitz have stated that it is fairly certain that trauma to an intradermal nevus will not induce malignancy. Whether simple trauma to a junctional nevus will produce malignancy is not known.

On the other hand, it is now well recognized that the congenital nevi have a great risk for transformation to malignant melanoma. Kaplan reported seven cases which were seen at the Stanford University Hospital and discussed 49 other cases. It is estimated that 14% of the large congenital nevi may undergo malignant transformation. Of particular interest is the occurrence of the B-K mole syndrome described by Clark and coworkers. This autosomal dominant condition is characterized by large pigmented nevi and a high risk for the development of melanoma. Its occurrence; however, has not been documented intraorally.

Surgical excision of all intraoral pigmented nevi is recommended as a prophylactic measure because of the constant chronic irritation of the mucosa in nearly all intraoral sites occasioned by eating, toothbrushing, etc.

Definable White Lesions

Examples include:

Premalignant/Potentially Malignant Lesion

In two studies from India (Gupta PC et al, 1980, Silverman S et al, 1976), rather low annual malignant transformation rates of oral leukoplakia have been reported, 0.3% and 0.06%, respectively. In reports from western countries, usually based on hospital material, somewhat higher figures have been mentioned (Axell T, 1987). One must take the following into account, when studying percentages of malignant transformation rates of oral leukoplakia:

On the basis of the lowest reported annual malignant transformation rate of oral leukoplakia, it can be calculated that patients with oral leukoplakia carry a five fold higher risk of developing oral cancer than controls. Recently the role(s) of dietary factors in determining the precancerous nature of oral leukoplakia among tobacco habitues revealed interesting results (Gupta PC et al, 1998). A population-based case control study in the Bhavnagar district, Gujarat, India estimated nutrient intake in blinded, house-to-house interviews. Among 5,018 male tobacco users, 318 were diagnosed as cases. An equal number of controls matched on age (± 5 years), gender, village, and use of tobacco were selected. Odd ratio (OR), a measure of the relative risk of malignant transformation of individual lesions were calculated from multiple logistic regression analysis controlling for relevant variables (type of tobacco use and economic status), a protective effect of fiber was observed for both oral submucous fibrosis (OSF) and leukoplakia; with 10% reduction in risk per gm day (p<0.05). Ascorbic acid (vitamin C) appeared to be protective against leukoplakia with the halving of associated risk of malignant transformation (OR 0.46; p<0.10). It is apparent that in addition to tobacco use, intake of specific nutrients and their deficiency may have a role in the development and progression of oral precancerous lesions.

Epidemiology

The prevalence of this lesion in Ernakulam district (Kerala, India) was 17 per 1,000; it was highest (61 per 1,000) among people with mixed habits. The annual age adjusted incidence rate was 2.1 per 1,000 among men and 1.3 per 1,000 among women; the highest incidence (6.0 per 1,000) was among men who both chewed and smoked. In an adult Swedish population a 3.6% prevalence rate was recorded (Axell T, 1976). Almost all leukoplakia in India occur in tobacco users. A definite dose-response relationship between leukoplakia and various forms of tobacco use in this area has been demonstrated. The dose-response relationship was stronger for the smoking habit than for the chewing habit and remained significant after taking account of age, gender and type of tobacco habit.

Age and Gender

The onset of leukoplakia usually takes place after the age of 30 years; resulting in a peak incidence above the age of 50 years. The gender distribution in most studies varies, ranging from a strong male predominance in different parts in India, to almost 1 : 1 in the western world.

Etiology

Leukoplakia occurs more frequently in smokers of tobacco than in nonsmokers. There is a dose-response relationship between tobacco usage and the prevalence of oral leukoplakia. Reducing or cessation of tobacco use may result in the regression or disappearance of oral leukoplakia (Gupta et al, 1995). On the other hand, disappearance of oral leukoplakia has occasionally been reported in patients who continued to smoke (Silverman and Rozen, 1968). Tobacco was most often chewed as an ingredient in betel quid (smokeless tobacco or paan) in India. The paan-chewers’ lesion consists of a thick, brownish-black encrustation on the buccal mucosa at the site of the placement of betel quid. It is often seen in heavily addicted betel quid chewers. It could be scraped off with a piece of gauze (leukoplakia); it regresses spontaneously when the habit is discontinued. Due to these reasons the paan-chewer’s lesion does not deserve the designation of leukoplakia. This is a specific entity and rarely progresses to leukoplakia. Whether the use of alcohol by itself is an independent etiological factor in the development of oral leukoplakia, is still questionable. Its effect, at best, may be synergistic to other well-known etiological factors (physical irritants). The role of Candida albicans as a possible etiological factor in leukoplakia and its possible role in malignant transformation is still unclear. About 10% of oral leukoplakias satisfy the clinical and histological criteria for chronic hyperplastic candidiasis (candidal leukoplakia). Epithelial dysplasia is reported to occur four to five times more frequently in Candida leukoplakia than in leukoplakia in general. However, this change is more common in the speckled variant than in homogeneous leukoplakia and carcinomatous change is more a characteristic of the speckled lesion than that of candidal superinfection. Various kinds of evidence has been presented to justify an etiologic role for candida in neoplastic transformation, which includes, among others, the catalytic transformation in vitro of the carcinogenic nitrosamine, N-nitrosobenzyl-methylamine, by strains of C. albicans demonstrated to be selectively associated with leukoplakia. The possible contributory role of viral agents (human papilloma virus strains 16, 18) in the pathogenesis of oral leukoplakia has also been discussed, particularly with regard to exophytic verrucous leukoplakia (Palefsky JM et al, 1995). In a study from India, serum levels of vitamin A, B₁₂, C, beta-carotene and folic acid were significantly decreased in patients with oral leukoplakia compared to controls, whereas, serum vitamin E was not (Ramaswamy G et al, 1996). Relatively little is yet known with regard to possible genetic factors in the development of oral leukoplakia.

Clinical Aspects

Preleukoplakia is defined as a low-grade or very mild reaction of the oral mucosa, appearing as a gray or grayish-white, but never completely white area with a slightly lobular pattern and with indistinct borders blending into the adjacent normal mucosa (Pindborg et al, 1968).

Clinical Classification

It is desirable to record, separately the various forms of leukoplakia and for this purpose subdivisions are recommended (WHO, 1980) (Table 2-2).

The adjective ‘nonhomogeneous’ is applicable both to the aspect of color, i.e. mixture of white and red changes (erythroleukoplakia) and to the aspect of texture, i.e. exophytic, papillary or verrucous. With regard to the latter lesions, no reproducible clinical criteria can be provided to distinguish (proliferative) verrucous leukoplakia from the clinical aspect of verrucous hyperplasia or verrucous carcinoma. The homogeneous type is usually otherwise asymptomatic, whereas the nonhomogeneous (mixed white and red) leukoplakia are often associated with mild complaints of localized pain or discomfort. In the presence of redness or palpable induration, malignancy may already be present (Fig. 2-7).

PVL

First described by Hansen et al, in 1985, PVL continues to be recognized as a particularly aggressive form of oral idiopathic leukoplakia that has a considerable morbidity and a strong potential for malignant transformation. Diagnosis is often made late in the protracted course of PVL with the disease in an advanced stage when it is especially refractory to treatment. The histologic spectrum that is seen in PVL are:

VH

This is a forerunner of verrucous carcinoma and the transition is so consistent that the hyperplasia, once diagnosed, should be treated like verrucous carcinoma.

Histopathological Aspects

It should be emphasized that leukoplakia is a clinical term, and its use carries no implications with regard to the histological findings. However, it is recommended that a histological report should always include a statement on the presence or absence of epithelial dysplasia, and if present, the assessment of its severity. The hallmarks of the histopathological aspects of leukoplakia are epithelial hyperplasia and surface hyperkeratosis. Epithelial dysplasia, if present, may range from mild to severe. In some instances, carcinoma in situ and even squamous cell carcinoma are encountered histologically. The various cellular changes that may occur in epithelial dysplasia are listed in Table 2-3. The clinical significance of human papilloma virus-associated epithelial dysplasia, so-called ‘koilocytic dysplasia’ remains to be investigated. The term ‘lichenoid dysplasia’ is sometimes used when the dysplastic epithelium may show features that, to some extent, resemble those of lichen planus. The term ‘chevron’ type of keratinization is used when it is associated with use of tobacco. Microabscesses may be observed in the superficial layers of the epithelium in the presence of C. albicans and inflammatory cell infiltration is commonly seen. Some of the exophytic, verrucous or papillomatous lesions, in spite of the absence of epithelial dysplasia, may in time progress to squamous cell carcinoma and that long-term follow-up should be considered. The final diagnosis of a white lesion of the oral mucosa can often only be made through a close dialogue between the clinician and the pathologist. Even then, cases may remain unsettled.

Grading of Epithelial Dysplasia

Epithelial dysplasia is usually assessed subjectively and considerable interobserver variability exists in its interpretation. Principally, what is lacking in this exercise is objectivity and lack of reproducibility. A consensus decision is, by and large, adopted in the management of individual lesions and based on the presence of dysplastic features, epithelial dyplasia is usually divided into three categories: mild, moderate and severe. It is recommended that the histological report of a leukoplakia should include a statement on the absence or presence of epithelial dysplasia and an assessment of its severity. The practical value of the grading of epithelial dysplasia is questionable. Although leukoplakia with moderate or severe epithelial dysplasia shows a greater disposition for malignant transformation than in the absence of dysplastic features, carcinomatous transformation may also take place in non-dysplastic leukoplakias.

Elimination of Possible Cause(s)

The clinician should first try to rule out any of the definable white lesions before accepting a definitive clinical diagnosis of leukoplakia.

Biopsy

In homogeneous leukoplakia, the value of histological examination might, to some extent, be questioned. The occurrence of epithelial dysplasia is rather low in this clinical subtype, as is the risk of future malignant transformation. However, this cannot be taken as the dictum in all cases since at least in few cases it proves to be otherwise. Therefore, the taking of a biopsy in homogeneous leukoplakia should be the standard rule. If treatment consists of CO₂-laser evaporation, it is mandatory to have a biopsy taken prior to such treatment. In nonhomogeneous leukoplakia, biopsy should be taken at the site of symptoms, if present, and/or at a site of redness or induration. Diagnostic methods other than histological examination, such as the use of toluidine blue staining or Lugol’s iodine and exfoliative cytology are of limited value when dealing with leukoplakia.

Treatment Modalities

Apart from the surgical excision, various treatment modalities are available, such as cryosurgery, CO₂-laser surgery, retinoids and other drugs, and recently, photodynamic therapy. The last is a rather recent application with regard to oral leukoplakia, therefore does not deserve comment on long-term results.

Possible Solutions

This overview demonstrates that to-bacco smoking and betel quid chewing are detrimental to oral health, as they are strongly associated with oral leukoplakia, oral cancer and other mucosal pathologies. In view of these findings, specific studies for primary and secondary preven-tion of these lesions are warranted. Prevention was found to be feasible and effective (Gupta et al, 1980) by timely diag-nosis and management of oral precancerous lesions like leu-koplakia and by measures like habit alleviation. Effort in this direction will take us a long way into the overall control and management of potentially malignant oral mucosal lesions.

Clinical Features

The gross appearance of leukoedema varies from a filmy opalescence of the mucosa in the early stages to a more definite grayish-white cast with a coarsely wrinkled surface in the later stages (Fig. 2-8 A). Lesions occur bilaterally in the majority of cases and frequently involve most of the buccal mucosa, extending onto the oral surface of the lips. Leukoedema is most noticeable along the occlusal line in the bicuspid and molar region. In some cases, desquamation occurs, leaving the surface eroded.

Etiology

The etiology of leukoedema is unknown. An extensive study of this condition was carried out in the United States by Sandstead and Lowe, who found no apparent correlation between the incidence of the condition and the use of tobacco, the pH of the saliva, oral bacterial infection, syphilis or galvanic irritation. In their study, the incidence of leukoedema was approximately 45% in Caucasian men and 40% in Caucasian women, while it was 94% in Negro men and 86% in Negro women, all adults with an average age of 45 years.

It has now been identified in many countries around the world with a remarkable variation in prevalence that suggests some ethnic association. The different reported epidemiological studies have been reviewed by Axell and Henricsson, who also noted an incidence of occurrence of nearly 50% in a survey of over 20,000 persons aged 15 years and above in Sweden. Their data suggested that tobacco smoking was closely correlated with the occurrence of the leukoedema, although studies from India, where tobacco use is extremely prevalent, have shown a very low incidence of leukoedema.

Histologic Features

The microscopic features of leukoede-ma consist of an increase in thickness of the epithelium, intra-cellular edema of the spinous or malphigian layer, a superfi-cial parakeratotic layer of several cells in thickness, and broad rete pegs which appear irregularly elongated (Fig. 2-8 B). The characteristic edematous cells appear extremely large and pale, and they present a reticular pattern. The cytoplasm appears lost, and the nuclei appear absent, clear or pyknotic. Infla-mmatory cell infiltration of the subjacent connective tissue is not a common finding.

In a review and discussion of leukoedema, Archard and his coworkers have concluded that the clinical appearance of the lesion is due to a retained superficial layer of parakeratotic cells.

Clinical Significance

It has been suggested that leuko-edema may represent a lesion of the oral mucosa in which leukoplakia is more apt to develop than in normal epithelium. This conclusion is based upon the fact that nearly all patients examined who manifested leukoplakia also exhibited leuko-edema in the adjacent mucosa. However, Archard and his coworkers have concluded, on the basis of their studies, that there is no evidence to support this contention. Furthermore, they concluded that there is no evidence that this lesion is pre-malignant or in any way associated with potential malignant alteration.

Since leukoedema appears to be simply a variant of normal mucosa, no treatment is necessary.

Clinical Features

In a study of 82 lesions of carcinoma in situ in 77 patients reported by Shafer, he found that the clinical appearance was that of a leukoplakia in 45% of the lesions, an erythroplakia in 16% of the lesions, a combination of leukoplakia and erythroplakia in 9%, an ulcerated lesion in 13%, a white and ulcerated lesion in 5%, a red and ulcerated lesion in 1% and not stated in 11%.

These lesions have been reported to occur at all intraoral sites but the most common in the study of Shafer were floor of mouth (23%), tongue (22%) and lips in males (20%). This distribution is roughly comparable to that found in a series of 158 lesions of asymptomatic early oral epidermoid carcinoma reported by Mashberg and his associates. They appear to be somewhat more common in men than in women (1.8 : 1) and tend to occur principally in elderly persons.

Histologic Features

Intraepithelial carcinoma is character-ized by a remarkable range of variation in histologic appear-ance. Keratin may or may not be found on the surface of the lesion but, if present, is more apt to be parakeratin rather than orthokeratin. Individual cell keratinization and epithelial or keratin pearl formation are exceedingly rare. In fact, these ap-pear to be a hallmark of transformation of carcinoma in situ into invasive carcinoma, so that if these are found, a further search should be made for carcinomatous invasion. In some instances, there appears to be hyperplasia of the altered epithe-lium, while in others there is atrophy.

Certain cytologic alterations may also occur. An increased nuclear/cytoplasmic ratio and nuclear hyperchromatism are sometimes seen, but many cases do not show these. Cellular pleomorphism is quite uncommon. One of the most conspicuous and consistent alterations is loss of orientation of cells and loss of their normal polarity. Mitotic activity is extremely variable and of little significance unless overwhelming (Fig. 2-9).

It has also been found that sometimes a sharp line of division between normal and altered epithelium extends from the surface down to the connective tissue rather than a blending of epithelial changes. It is also not unusual to find multiple areas of carcinoma in situ interspersed by essentially normal appearing epithelium, producing multifocal carcinoma in situ.

All of the above changes occur within the surface epithelium, which remains confined by the basement membrane.

Treatment and Prognosis

There is no uniformly accepted treatment for intraepithelial carcinoma. The lesions have been surgically excised, irradiated, cauterized and even exposed to solid carbon dioxide. If the condition is left untreated, carcinomatous invasion is thought to occur eventually. Spontaneous regression of carcinoma in situ without treatment is known to occur in certain sites, chiefly the uterine cervix, in a significant percentage of cases. However, it is doubtful that this happens in the oral cavity; although progression into invasive carcinoma may take years in some instances, in others it apparently develops within months.

Histopathologic Features

The epithelium shows lack of keratin production and is often atrophic, but it may be hyperplastic. This lack of keratinization, especially when combined with epithelial thinness, allows the underlying microvasculature to show through, thereby causing the red color. The underlying connective tissue often demonstrates chronic inflammation. Differentiation of erythroplakia with malignant change and other early squamous cell carcinomas from benign inflammatory lesions of the oral mucosa can be enhanced by use of 1% toluidine blue (tolonium chloride) solution applied topically with a swab or as an oral rinse. This technique gives excellent results in detecting epithelial dysplasia with false-negative (underdiagnosis) and false positive (overdiagnosis) rates of well below 10%.

Treatment

It should follow the same principles outlined for leukoplakia. Observation for one to two weeks following elimination of suspected irritants is acceptable. Prompt biopsy is thereafter mandatory for lesions that persist. Surgical excision gives excellent results and a recurrence rate of less than 5% is reported.

Epidemiology

The prevalence of this lesion among Indians was 0.3%; 52% of the 31 lesions occurred among bidi smokers (Mehta FS et al, 1971). The annual age adjusted incidence rate among smokers was 1.7 per 1,000; 0.7 per 1,000 in those who smoked and chewed (Gupta PC et al, 1980).

Natural History

Over a longer period 66% of the 44 lesions remained stationary, 34% regressed spontaneously and none showed malignant transformation. Palatal changes in reverse smokers (smoke with the burning end of a cigar/ cigarette within the oral cavity), on the other hand, must be distinguished from this lesion and are multimorphic and precancerous, whereas leukokeratosis nicotina palati exhibits neither great variability nor malignant transformation (Table 2-4).

Epidemiology

The annual age-adjusted incidence rates of palatal changes (encompassing all components) was 24.9 per 1,000 among men and 39.6 per 1,000 among women and the peak incidence was in the 55–64 year age group (Srikakulam data).

Clinical Aspect

Palatal changes comprise several compo-nents:

Histological Features

Hyperorthokeratosis, epithelial dys-plasia and inflammatory cells in the connective tissue were observed in 87%, 23% and 55% of the biopsies collected from Srikakulam district (Daftary et al, 1992). Melanin depos-its were noted in the lamina propria of most of them. The epithelium was atrophic in 60% of biopsies from red areas. Epithelial dysplasia was observed in 52% of red areas, 25% of excrescences, 20% of ulcerations, 10% of patches and in 19% of nonpigmented areas.

Natural History

In a six-year follow-up study, palatal changes remained stationary in 75% of individuals; regressed in 14% and were variable in 11%, i.e. they regressed, recurred and regressed again (Gupta PC et al, 1980). The regression rates were higher when the habit was discontinued or reduced substantially. Malignant transformation was observed for 0.3% of the palatal lesions. In a 10-year intervention study in the same area, the malignant potential of various components of palatal changes was evaluated: red areas and patches were found to exhibit a high potential for malignant transformation.

Epidemiology

Of the 69 lesions observed among 7,216 tobacco users, 87% occurred among smokers, especially bidi smokers.

Clinical Aspects

This lesion occurs either independently or sometimes with other lesions. About 10% of the lesions were associated with palatal papillary hyperplasia and 25% with central papillary atrophy of the tongue and bilateral commissural leukoplakias. This triad of lesions is comparable to the multifocal candidiasis described in western literature.

Natural History

The highest percentage of persistent lesions (60%) was seen among people who did not give up their smoking habits, while the highest percentage (75%) of regressed lesions occurred among those who discontinued or reduced smoking substantially. These observations clearly indicate that palatal erythema is caused by smoking, particularly bidi smoking. None of the lesions progressed to cancer.

Epidemiology

The prevalence of this lesion in the general population was 1%; it was present among 2.2% bidi smokers, 1.6% cigarette smokers and 0.3% of nonsmokers of tobacco. In a 10-year follow-up study from India, the annual age-adjusted incidence rate among smokers was 1.5 per 1000 as compared to 0.8 per 1,000 among nonsmokers.

Etiology

It is considered to be due to candidal infection, smoking or both (bidi smoking). Interestingly, very few women had this lesion due perhaps to the rarity of smoking among women.

Histologic Features

This lesion was marked by the absence of tongue papillae, the presence of slight parakeratinization of the epithelial surface, long slender rete ridges and occasional pseudoepitheliomatous hyperplasia. Chronic inflammatory cell infiltrate, chiefly of lymphocytes, was usually present within the epithelium and in the lamina propria. Candidal hyphae were observed in the superficial layers of epithelium in the majority of cases.

Natural History

The highest percentage of regressed lesions (87%) was seen in those who stopped their habits. None of the lesions progressed to cancer.

Epidemiology

The annual age-adjusted incidence rate of this lesion was 28 per 1,000 among male chewers and 17.4 per 1,000 among female chewers (Gupta PC et al, 1980).

Histologic Features

These lesions showed pale staining parakeratin-like surface layers of epithelium, containing round nuclear remnants, ballooning and vacuolated cells and epithelial hyperplasia.

Natural History

It is a specific entity and rarely progresses to leukoplakia. Malignant transformation was not observed in these lesions.

Epidemiology

The prevalence of this lesion among 5,099 individuals in Ernakulam, Kerala was 0.7%. About 89% of the lesions occurred among betel quid chewers and 11% among people with mixed habits. The annual age-adjusted incidence rates among men and women were 0.7 and 2.2 per 1,000 respectively. The peak incidence for women was in the 45–54 year age group. The incidence was zero among smokers and nonusers of tobacco; and 4.3 per 1,000 among women who chewed. This lesion is thus entirely associated with betel quid chewing.

Clinical Aspects

The lesion always occurs on the buccal mucosa and mandibular groove, areas which are in intimate contact with the betel quid.

Histologic Features

The lesion shows parakeratinized, atrophic epithelium, liquefaction degeneration of the basal cell layer and a band-like inflammatory cell infiltrate containing lymphocytes and plasma cells. Unlike lichen planus, this lesion shows hyperparakeratosis, and plasma cells in the juxtaepithelial region.

Natural History

Although the histological features are similar to those of oral lichen planus, in view of its complete association with betel quid chewing, it is regarded as a specific entity. No malignant transformation has been reported.

Clinical Features

The onset is insidious, over a two to five years.

Prodromal Symptoms (Early OSF)

This includes a burning sensation in the mouth when consuming spicy food, appearance of blisters especially on the palate, ulcerations or recurrent generalized inflammation of the oral mucosa, excessive salivation, defective gustatory sensation and dryness of the mouth. There are periods of exacerbation manifested by the appearance of small vesicles in the cheek and palate. The intervals between such exacerbation vary from three months to one year. Focal vascular dilatations manifest clinically as petechiae in the early stages of the disease. This may be part of a vascular response due to hypersensitivity of the oral mucosa towards some external irritant like arecanut products. Petechiae are observed in about 22% of OSF cases (Rajendran, 1994), mostly on the tongue followed by the labial and buccal mucosa with no sign of blood dyscrasias or systemic disorders. Pain in areas where submucosal fibrotic bands are developing when palpated is a useful clinical test. Histologically, they revealed a slightly hyperplastic epithelium, sometimes atrophic with numerous dilated and blood-filled capillaries juxtaepithelially (Figs. 2-11, 2-12). The inflammatory cells seen are mainly lymphocytes, plasma cells and occasional eosinophils. The presence together of large numbers of lymphocytes and fibroblasts as well as plasma cells in moderate numbers, suggests the importance of a sustained lymphocytic infiltration in the maintenance of the tissue reaction in OSF.

Advanced OSF

As the disease progresses, the oral mucosa becomes blanched and slightly opaque, and white fibrous bands appear. The buccal mucosa and lips may be affected at an early stage, although it was thought that the palate and the faucial pillars are the areas involved first. The oral mucosa is involved symmetrically (with possible exception) and the fibrous bands in the buccal mucosa run in a vertical direction. The density of the fibrous deposit varies from a slight whitish area on the soft palate, causing no symptoms, to a dense fibrosis, causing fixation and shortening or even deviation of the uvula and soft palate. The fibrous tissue in the faucial pillars ranges from a slight submucosal accumulation in both pillars to a dense fibrosis extending deep into the pillars with strangulation of the tonsils. It is this dense fibrosis, involving the tissues around the pterygomandibular raphae, that causes varying degrees of difficulty in mouth opening. A factor which seems to be overlooked by many investigators while recording the extent of mouth opening is the acuteness of oral symptoms (persistent/recurrent glossitis and stomatitis) at the time of recording. Sometimes the fibrosis spreads to the pharynx and down to the piriform fossae. Upon palpation, a circular band can be felt around the entire rima oris (mouth orifice), and these changes are quite marked in the lower lip. All observers have noted impairment of tongue movement in patients with advanced OSF with significant atrophy of the tongue papillae. With progressing fibrosis, the stiffening of certain areas of the mucosa occurs leading to difficulty in opening the mouth, inability to whistle or blow out a candle and difficulty in swallowing. When the fibrosis involves the nasopharynx, the patient may experience referred pain to the ear and a nasal voice as one of the later signs in some patients.

Epidemiology

Numerous published reports on OSF allow an informed appraisal of its geographic distribution, together with data on the percentage prevalence. An epidemiological assessment of the prevalence of OSF among Indian villagers, based on baseline data, recorded a prevalence of 0.2% (n 10,071) in Gujarat, 0.4% (n 10,287) in Kerala, 0.04% (n 10,169) in Andhra Pradesh, and 0.07% (n 20,388) in Bihar. The prevalence among 101,761 villagers in the state of Maharashtra (central India) was 0.03%. In a 10-year follow-up study of oral precancer, Gupta et al, in 1980, calculated the incidence rate of OSF in Ernakulam, Kerala: 8 for men and 19 for women per 100,000. Variations in the prevalence figures are common between different studies, probably because of differences in the clinical criteria for diagnosis. While some investigators adhere to the earlier signs and symptoms, others looked for fibrous bands as the diagnostic criterion. According to Pindborg 1989, if only the fibrous band was the criterion for diagnoses, the prevalence rate would have been about 1.6%. Prevalence by gender varies widely in the different published studies. The general female preponderance may be related to factors like oral habits, deficiency states of iron, vitamin B complex among many other conditions prevalent in Indian women.

Etiology

There is compelling evidence to implicate the habitual chewing of arecanut with the development of OSF. It occurs predominantly in the Indian subcontinent where the habit is more prevalent. The frequency of this habit in population affected by OSF ranged from 34–100% (Bhonsle RB et al, 1987). This has been reported to be higher among OSF patients than in the general population. In a study of 100,000 villagers in Maharashtra (India), 4.2% of females who chewed areca nut and did not use tobacco suffered from OSF. Thus chewing areca nut may be an important factor in the etiology of OSF. Previous studies on the pathogenesis of OSF have suggested that the occurrence may be due to:

Genetic susceptibility may also be associated with OSF because raised frequencies of HLA-A10, – B7 and – DR3 are found in OSF patients compared to normal subjects. Further HLA-typing done by the use of the polymerase chain reaction (PCR) also demonstrates significantly increased frequencies of HLA-A24, DRB1-11 and DRB3-0202/3 antigens in 21 OSF patients when compared with the English controls (Saeed B et al, 1997).

Management

The reduction or even elimination of the habit of areca nut chewing is an important preventive measure. At least in the early stages of OSF, it could probably slow the progress of the disease. To improve current treatment regimens of OSF, the following strategies have been proposed:

Etiology

Ultraviolet radiation is the most important and common cause of BCC. Shorter wavelength ultraviolet (UV) radiation (290–320 nm, sunburn rays) is believed to play a greater role than the longer wavelength ultraviolet A (UVA) radiation (320–400 nm, tanning rays). Thus chronic sun exposure is important in the development of BCC. A long latency period of 20–50 years is typical between the time of UV damage and clinical onset of BCC. X-ray exposure as well as chemical like arsenic has been associated with basal cell carcinoma development. A modest role has been ascribed to immunosuppression as well.

Syndromes like xeroderma pigmentosum (due to an inability to repair UV-induced DNA damage), and nevoid BCC syndrome are characterized by multiple basal cell carcinomas occurring in early age.

Basal cell carcinoma is believed to arise from the pluripotential stem cell compartments of the basal layer of epidermis as well as follicular structures (hair follicle stem cells residing just below the sebaceous gland duct).

Clinical Features

Basal cell carcinoma occurs most fre-quently in persons in the fourth decade of life or later, but it has been reported as occurring in younger persons, even children. The male-to-female ratio is approximately 3:2. It is a disorder of white individuals, especially those with very fair skin. It is rare in individuals with dark skin.

Basal cell carcinoma is most frequently seen in the middle third of the face, but may occur anywhere on the sun-exposed part of skin. It is important to realize that this form of carcinoma does not arise from oral mucosa and thus is never seen in the oral cavity unless it arrives there by invasion and infiltration from a skin surface.

The subtypes of basal cell carcinoma are:

Histologic Features

Tumor cells of nodular basal cell carcinoma, sometimes called basalioma cells, typically have large, hyperchromatic, oval nuclei with little cytoplasm and arranged in nests of varying size. Cells appear rather uniform, and mitotic figures are usually few in number. The cells at the periphery of these nests show palisading. Early lesions usually exhibit some connection to the overlying epidermis, but such contiguity may be difficult to appreciate in more advanced lesions. Increased mucin is often present in the surrounding dermal stroma.

Cleft formation, known as retraction artifact, commonly occurs between BCC nests and stroma because of shrinkage of mucin during tissue fixation and staining. Some lobules may exhibit areas of pseudoglandular change, and this is the predominant change in adenoid basal cell carcinoma. In the nodulocystic variant, larger tumor lobules may degenerate centrally, forming pseudocystic spaces filled with mucinous debris.

In pigmented basal cell carcinoma, benign melanocytes in and around the tumor produce large amounts of melanin. These melanocytes contain many melanin granules in their cytoplasm and dendrites.

Superficial basal cell carcinoma appears as buds of basaloid cells attached to the undersurface of the epidermis. Nests of various sizes often are seen in the upper dermis which show typical palisading periphery.

The more aggressive morpheaform and infiltrating basal cell carcinomas exhibit growth patterns resulting in strands of cells rather than round nests. Morpheaform BCC arises as thin strands of tumor cells that are embedded in a dense fibrous stroma. The strands of infiltrating type tend to be somewhat thicker than those seen in morpheaform basal cell carcinoma, and have a spiky irregular appearance. Infiltrating basal cell carcinoma usually does not exhibit the scar like stroma seen in morpheaform type. Peripheral palisading and retraction are much less pronounced in morpheaform and infiltrating basal cell carcinoma and subclinical involvement is often extensive.

Another aggressive variant, micronodular basal cell carcinoma, appears as small nodular aggregates of basaloid cells. Retraction artifact tends to be less pronounced than in the nodular form and subclinical involvement is often significant.

Treatment and Prognosis

Each lesion must be considered separately when contemplating the choice of therapy. In general, equally good results may be expected from surgical excision of the tumor or from X-ray radiation. There is probably an equal number of failures or recurrences subsequent to each type of treatment. Some lesions stubbornly resist treatment and exhibit great propensity for recurrence and subsequent destruction of tissue.

The prognosis of basal cell lesions is good, since the neoplasm grows slowly, does not tend to metastasize and responds well to treatment. Most difficulties, which may lead to death by local invasion, are due to neglect on the part of the patient who fails to seek medical aid until the lesion is in a far-advanced state.

Epidemiology

The incidence of squamous cell carcinomas of the oral cavity differs widely in various parts of the world and ranges from approximately 2–10 per 100,000 population per year. Such differences can, to some extent, be explained on the basis of environmental differences or lifestyle and habits among certain populations, such as betel quid chewing, snuff dipping or the habit of reverse smoking. High incidence countries include those in South Asia such as Sri Lanka, India, Pakistan and Bangladesh; Bas-Rhin and Calvados regions in France; countries in central and eastern Europe; and Brazil. The incidence of oral carcinoma in blacks is somewhat lower than in whites, which is mainly due to the low incidence of lower lip cancer in blacks.

In most parts of the world the male-female ratio is approximately 2 : 1 for oral carcinoma, except for carcinoma of the vermilion border of the lower lip. In the latter site there is a strong male predominance. Oral squamous cell carcinomas are mainly found after the fourth decade.

Etiology

The use of tobacco in its various forms, including smokeless tobacco, is regarded as the main cause of oral cancer, particularly when associated with the use of excess alcohol. High exposure to ultraviolet light increases the chance of developing cancer of the lower lip. Diets with low levels of vitamins A and C or inadequate consumption of vegetables and fruits may contribute to the risk of oral cancer.

Patients who are immunosuppressed, e.g. renal and homograft recipients and HIV-infected patients, have a higher incidence of subsequent cancer development, particularly of the lower lip. Furthermore, a number of rare conditions predispose to the development of oral cancer, such as xeroderma pigmentosum, Fanconi’s anemia, and Bloom’s syndrome.

In some patients with oral cancer, especially in females, none of the aforementioned factors or cofactors seem to be present, which makes the etiology of the development of oral cancer incompletely understood.

Although much circumstantial evidence currently exists, the literature still lacks unequivocal evidence linking HPV and oral cancer. When all anatomic sites are considered, the prevalence of HPV in oral cancer does not appear to be higher than the prevalence of HPV in clinically normal mucosa.

Epidemiological findings also lend further support to a causal relationship between HPV and oral cancer, particularly for oropharyngeal tumors:

However, it is difficult to ignore the highly preferential association of HPV with tonsillar carcinoma, as well as the finding of high-risk HPV type 16 in these tumors. It is also possible that this area is more predisposed to epithelial injury which facilitates the subsequent viral infection.

Sunlight, in cases of lip cancer, appeared to be of only minor etiologic significance. This is the generally accepted reason for its occurrence preponderantly in fair-skinned outdoor male workers who are affected by exposure of the lower lip to the sun, whereas the upper lip is partially protected. The lesser rate of occurrence in females may be due to lower exposure to sunlight and the partial protection of their lips by lipstick.

Trauma and dental irritation were not found to be significant etiologic factors in oral cancer. In this regard, Monkman and his associates have also carried out a thorough review of the literature relating cancer to trauma and found no evidence to suggest that single uncomplicated trauma can cause cancer. They also found it debatable whether trauma can aggravate or accelerate existing malignancies. However, they concluded that trauma, in combination with other factors, may act as a cocarcinogen and that there was adequate evidence suggesting that metastatic spread of malignant tumors can be affected by trauma.

Clinical Presentations of Oral Cancer

Clinically, almost all oral cancers, except those in the earliest stages have two very characteristic features in the form of ulceration and an indurated margin. In different sites; however, there are certain variations.

Histologic Features

Considerable histologic variation is presented in intraoral epidermoid carcinomas, although in general they tend to be moderately well-differentiated neoplasms with some evidence of keratinization (Fig. 2-21 and Table 2-9). Highly anaplastic lesions do occur, but are uncommon; such lesions tend to metastasize early and widely and cause death quickly. The well-differentiated epidermoid carcinoma consists of sheets and nests of cells with obvious origin from squamous epithelium. These cells are generally large and show a distinct cell membrane, although intercellular bridges or tonofibrils often cannot be demonstrated. The nuclei of the neoplastic cells are large and may demonstrate a good deal of variability in the intensity of the staining reaction. Nuclei which stain heavily with hematoxylin are referred to as hyperchromatic. In the well-differentiated lesion mitotic figures may be found, but they often do not appear to be especially numerous (Fig. 2-22). Many of these mitotic figures are atypical, although this may be obvious only to an experienced histopathologist. One of the most prominent features of the well-differentiated epidermoid carcinoma is the presence of individual cell keratinization and the formation of numerous epithelial, or keratin pearls of varying sizes. In a typical lesion, groups of these malignant cells can be found actively invading the connective tissue in a vagarious pattern.

Less well-differentiated epidermoid carcinomas lose certain features, so that their resemblance to squamous epithelium is less pronounced. The characteristic shape of the cells and their arrangement may be altered. The growth rate of the individual cells is more rapid, and this is reflected in the greater numbers of mitotic figures, the even greater variation in sizes, shape and tinctorial reaction and the failure to carry out the function of a differentiated squamous cell, the formation of keratin.

The poorly differentiated carcinomas bear little resemblance to their cell of origin and will often present diagnostic difficulties because of the primitive and uncharacteristic histologic appearance of the malignant, rapidly dividing cells. These cells show an even greater lack of cohesiveness and are extremely vagarious.

The recognition that different degrees of differentiation occur in the epidermoid carcinoma prompted Broders to suggest a system of grading tumors in which a grade I lesion was highly differentiated (its cells were producing much keratin), while grade IV was very poorly differentiated (the cells were highly anaplastic and showed practically no keratin formation). The fact that the same tumor may show different degrees of differentiation in varying areas has prompted the discontinuation of the grading system. Instead, most pathologists now modify the diagnosis of the neoplasm by a descriptive adjective indicative of its differentiation. The one advantage of grading a tumor is that the grade reflects the anaplasticity of the lesion, which in turn indicates the general rapidity of growth, the rapidity of metastatic spread, the general reaction to be expected after X-ray radiation and the prognosis.

This great advantage is still retained in the descriptive grading which is replacing the numerical grading system.

Metastases from intraoral carcinoma of different sites involve chiefly the submaxillary and superficial and deep cervical lymph nodes (Figs. 2-23-2-26–2-26). Occasionally, other nodes such as the submental, preauricular and postauricular nodes and supraclavicular nodes may be involved, but blood stream metastasis from oral cancer is uncommon. However, Gowen and de Suto-Nagy have reported on a series of 59 patients who died of carcinoma of the head and neck and were autopsied, with distant metastases being specifically sought. Surprisingly, 57% of these patients did have distant metastases; in the lung in 82% of the cases, in the liver in 45%, and in the bones in 23%.

1. External upper lip (vermilion border) (C00.0)

2. External lower lip (vermilion border) (C00.1)

3. Commissures (C00.6)

1. Buccal mucosa

2. Upper alveolus and gingiva (upper gum) (C03.0)

3. Lower alveolus and gingiva (lower gum) (C03.1)

4. Hard palate (C05.0)

5. Tongue

6. Floor of mouth (C04).

Regional Lymph Nodes

The regional lymph nodes are the cervical nodes.

Mild dysplasia (Grade-1)

Demonstrates proliferation or hyperplasia of cells of the basal and parabasal layers which does not extend beyond the lower third of the epithelium. Cytological atypia is generally slight with only mild pleomorphism of cells or nuclei. Mitoses are not prominent, and when present are usually basally located and normal. Architectural (gross) changes are minimal.

Moderate dysplasia (Grade-2)

Demonstrates a proliferation of atypical cells extending into the middle one third of the epithelium. The cytological changes are more severe than in mild dysplasia and changes such as hyperchromatism, and prominent cell and nuclear pleomorphism may be seen. Increased and abnormal mitoses may be seen, but these are usually located at the basal layers of the epithelium. Architectural changes may be seen in the lower half of the epithelium where there may be loss of basal polarity and hyperplasia leading to bulbous rete pegs. However, stratification and maturation are relatively normal, often with hyperkeratosis of the surface epithelium.

Severe dysplasia (Grade-3)

There is abnormal proliferation from the basal layer into the upper third of the epithelium Cytological and architectural changes can be very prominent All the changes seen in mild and moderate dysplasia are present but in addition there is marked pleomorphism often ith abnormal large nuclei and prominent or even multiple nucleoli present. Prominent and suprabasal mitoses are usually evident and abnormal tripolar or star shaped forms may be seen. Apoptotic bodies (cell death) may also be prominent. Architectural changes are severe, often with complete lack of stratification and with deep abnormal keratinization and even formation of keratin pearls. Abnormal forms of rete pegs are usual and bulbous rete pegs are particularly significant in the diagnosis of severe dysplasia. Abnormal shaped rete pegs may also be seen, with lateral extensions or small branches. These are quite abnormal and may be the earliest signs of invasion. Occasional lesions may show prominent acantholysis with severe disruption of the architecture. Although the epithelium may be thickened, severe dysplasia is sometimes accompanied by marked epithelial atrophy (thinning). This is especially prominent in lesions from the floor of the mouth, ventral tongue or soft palate and may be a feature of lesions which have presented clinically as erythroplakia (red patch). In these cases there may be minimal evidence of stratification or keratinization, and atypical cells may be extended to the surface.

Carcinoma in situ

It is the most severe form of epithelial dysplasia and is characterised by full thickness cytological and architectural changes. In the oral cavity such changes are rare, and often, even in the presence of the most severe atypia, there is still an intact keratinised surface layer. Carcinoma in situ is thought by some to be a premalignancy but others regard it as evidence of actual malignant change but without invasion (Speight PM, 2010).