Premalignancy
Premalignant Oral Mucosal Diseases
Raymond J. Melrose, DDS
Copyright 2001 Journal of the California Dental Association.
 |
A premalignant phase in the development of oral cancer is predicted by the classic model of experimental epithelial carcinogenesis. Virtually all oral squamous cell carcinomas arise from a premalignant precursor, but it is difficult to specifically define the term premalignant. Oral pathologists use the term epithelial dysplasia to indicate microscopic features in a biopsy specimen that are associated with a risk of malignant change and then assign a grade of severity. There is good correlation between higher grades of dysplasia and increasing risk of cancer but less so with the lower grades. The clinical appearances manifested by oral epithelial dysplasia and early oral cancer include leukoplakia, erythroplakia. and speckled leukoplakia. This paper discusses and illustrates these clinical lesions, their associated risk factors, their relationship to epithelial dysplasia, and the associated risk of evolution into oral cancer.
|
The clinical concept of oral mucosal premalignancy is now more than 150 years old. Sir James Paget is credited with the first description of an association between an oral lesion (which he termed "ichthyosis") and subsequent development of tongue carcinoma.1 In fact, Paget had recognized 19 years earlier that oral mucosal white patches in pipe smokers bore a risk of transforming into cancer. Seven years after Paget’s report, Schwimmer described white lesions of the tongue (which he termed "leukoplakia") that developed into cancer in patients with tertiary syphilis.2 Thus, the term "leukoplakia" and its association with oral mucosal premalignant disease has been extant in the literature for 128 years.
Similarly, a red lesion of the glans penis in a syphilitic man that evolved into cancer was designated "erythroplasia" by Queyrat.2 Darier, in 1924, was the first to describe an oral mucosal red lesion with the potential to turn into cancer and designated this the "erythroplasia of Queyrat."2
The biological basis for the concept of oral premalignancy has been established by studies of epithelial carcinogenesis. From the earliest studies in mouse and hamster models through today’s elegant molecular and gene research has come confirmation that human epithelial cancer is not the result of a single random event. Instead, it has been established that epithelial carcinogenesis requires multiple "hits" over time that result in accumulated changes allowing now abnormal cells to proliferate while resisting both programmed cell death (apoptosis) and the body’s own immune surveillance.
The pathologic process of carcinogenesis takes time as well as an appropriate environment in order to produce a clinically perceptible result. That both time and a proper environment are required in most instances explains why the people at greatest risk of oral cancer are older adults who habitually use products such as tobacco and alcohol that contain carcinogens or that act as co-carcinogens.
Based upon years of clinical observations and buttressed by increasingly sophisticated research, it can be fairly stated that virtually all oral squamous carcinomas evolve out of an existing premalignant lesion. But, do all premalignant lesions develop into squamous carcinomas? How does one define what a premalignant lesion is? There is no completely agreed upon answer to either question. What have evolved are attempts to assess the risk or the likelihood of progression into cancer.
Clinicians have developed complex schemes of varied clinical appearances that attempt to correlate surface features such as color and texture with anticipated histologic appearance and a corresponding degree of epithelial dysplasia or carcinoma.3,4 For example, smooth, homogenous leukoplakia is assessed as having virtually no potential to harbor dysplasia; granular or verruciform leukoplakia is assessed as likely to show moderate to severe dysplasia on biopsy; and speckled leukoplakias are predicted to harbor severe dysplasia or carcinoma in situ. Such attempts at clinical pathologic correlation have merit but are flawed based upon observer bias and experience. Further, such schemes may result in all homogeneous leukoplakias being dismissed as "clinically benign" regardless of other important clinical considerations and thus not biopsied for confirmation of the clinical diagnosis.
Oral pathologists look at patterns of cell growth, maturation, and nuclear characteristics in biopsy specimens and assign grades of severity to dysplastic changes observed. A problem extant in this method is subjectivity and lack of clear, uniformly applied criteria for the diagnosis of dysplasias. Consistent reproducibility in the microscopic interpretation of dysplasias even by experienced oral pathologists has been an unachievable goal.
A further and significant problem with assessing the potential for the real risk of a premalignant lesion evolving into a squamous carcinoma is the lack of an agreed upon, comprehensive staging system for oral premalignancy. Such a system has been proposed but has not yet been adopted.
A number of clinical studies of leukoplakias and their malignant transformation rates (the percent of patients with leukoplakias developing an associated cancer per year) have been reported.4 These data show a mean transformation rate of 8.1 percent with a variance from 3.6 percent to 19.8 percent. The wide variance between rates in reported studies is testimony to the inherent risk associated with applying such data to any one clinical situation.
Thankfully, researchers today are measuring detectable levels of tumor-suppressor genes such as p53, oncogenes and their protein products, epidermal growth factors, tumor cell angiogenesis, and a myriad of others in attempts to find objective, reproducible criteria. So far, no reliable predictor has appeared, but there is great promise in research attempting to elucidate a sequence or cascade of molecular changes that will correspond to rates of malignant change as they evolve in patients.
Even when reliable molecular markers of an evolving oral cancer are available, they will be useless without the clinician who first suspects a patient is at risk and who is looking for clinical alterations from normal. No universal screening test for oral cancer is on the horizon. Molecular techniques do require some form of a specimen as substrate. The test specimen might be as small as a few cells or may still require a biopsy. But either obtaining a few cells or selecting a biopsy site implies that the clinician is seeing the abnormality. Thus, despite remarkable advances that portend sophisticated ability to reliably differentiate dangerous lesions from those that are less dangerous, the bottom line today is that dentists are not relieved of the obligation to assess every patient’s risk for oral cancer, to look carefully for suspicious lesions, and to have these diagnosed as soon as possible after discovery.
Clinical Features
Oral mucosal premalignant lesions may manifest clinically in three principal forms: leukoplakia, erythroplakia, and speckled leukoplakia. Each of these will be discussed and illustrated.
Leukoplakia
Leukoplakia is defined as a "white patch that cannot be wiped off and for which no more-specific diagnosis can be given on clinical grounds." For example, neither mucosal candidiasis, which can be wiped off, nor reticular lichen planus, which has a specific clinical appearance, fits the definition.
Oral leukoplakias are common, most frequently innocuous, and related to low-grade local irritation that induces hyperkeratosis. Irritants most commonly associated with leukoplakias include mastication/parafunctional habits, removable prostheses, alcohol-containing products and beverages, and tobacco products. In assessing a given leukoplakia then, its relationship to local factors should be evaluated; and these should be eliminated to as great a degree as possible. Any leukoplakia for which there is no reasonable local cause, which fails to resolve with removal of presumed local irritants, or which is present in a high-risk site should be biopsied.
In a study of 3,360 biopsied leukoplakias, Shafer and Waldron found that 19.8 percent (670) were either mucosal dysplasia (premalignant) or squamous carcinoma.5 These authors also found that the floor of the mouth (42.9 percent), lateral tongue (24.2 percent), and soft palate (18.8 percent) had the highest rates of serious disease among leukoplakias. Therefore, in addition to a relationship to local irritants, the site of a leukoplakia is highly significant in a decision to biopsy.
Oral squamous carcinoma is uncommon before age 40, but the risk for this disease increases greatly after age 40. Since premalignant change precedes development of squamous carcinoma, it stands to reason that one should be even more suspicious of leukoplakias seen in patients older than 40.
The clinical appearances of leukoplakias vary. Those that are thin, homogeneous, and smooth (Figure 1) tend to be less worrisome, for example, than those that are thick and/or fissured (Figure 2). But, one should not rely on surface features as much as location, risk factors, and age in making the decision to biopsy.
Although leukoplakia is a nonspecific clinical descriptor and requires a biopsy to determine its nature, several specific conditions have been reported in recent years whose names include "leukoplakia." These are proliferative verrucous leukoplakia, oral hairy leukoplakia, and Viadent leukoplakia. Oral hairy leukoplakia is a disease induced by Epstein-Barr virus infection of epithelium in people with compromised immune systems such as in HIV infection.6 This condition is not premalignant and will not be discussed. Viadent leukoplakia is a unique condition in which keratosis develops in the anterior maxillary vestibule or attached mucosa presumably as a result of the sanguinaria incorporated into the product. Slight mucosal dysplasia has been reported in a few of these cases, but it is not known if these lesions have any malignant potential so they cannot be considered in a discussion of premalignant lesions at this time.7 Proliferative verrucous leukoplakia, on the other hand, is well-established as a premalignant process.8
Proliferative Verrucous Leukoplakia
Proliferative verrucous leukoplakia is an enigmatic condition first described by Hansen and colleagues in 1985. It cannot be diagnosed on the basis of a single biopsy. Instead, patients who will develop it present first with an idiopathic leukoplakia that, upon biopsy, is a benign keratosis or verrucous keratosis. These patients are usually older adults who have no risk factors for carcinoma. The gingiva is a commonly affected site along with the buccal mucosa and tongue (Figure 3). The biopsied lesion will recur and/or increase in size slowly over time. Often it will thicken and appear leathery or verrucous. Additional sites may appear as well. Repeat biopsies begin to show dysplasia; and this feature, in concert with the clinical appearance, results in a diagnosis of proliferative verrucous leukoplakia. In most instances, it will inexorably progress into frank squamous carcinoma or verrucous carcinoma (Figure 4). To date, no completely effective treatment has been found to eliminate proliferative verrucous leukoplakia. Reports of human papilloma virus strains in proliferative verrucous leukoplakia have been published; but antiviral therapy has not been effective, and it is not proved that it is a virus-induced disease. Topical chemotherapy as with Bleomycin has been attempted along with wide (nonradical) excision, but neither modality has been reliably effective. Use of a carbon dioxide laser for ablation has produced the longest disease-free intervals but still no documented cure.
Spit Tobacco Leukoplakia
A type of leukoplakia that has a specific etiology is that which arises secondary to chronic use of spit tobacco products, including chewing tobacco, moist snuff, and dry snuff. These leukoplakias, which may be thin or thick, arise at the site of placement of the product. Therefore, the buccal mucosa, mandibular vestibule, and buccal/facial gingiva are the sites most frequently affected. In some patients, the leukoplakia may develop within a few months after initiation of use, but it is more common for several years to pass before change is apparent. Often, the mucosal surface is wrinkled or fissured as well as white. This is because of a "pocket" that forms at the site where the quid of the product is placed. With the quid in place, the wrinkles disappear, leaving a greater absorptive surface for the nicotine and other products including carcinogens (Figure 5).
Those who habitually keep the quid in the same site have an increased risk of leukoplakia development compared with those who move the material from site to site. Snuff is said to be more likely to induce leukoplakias than chewing tobacco. Other factors that relate to increased incidence of leukoplakia are daily contact hours and the amount of product used daily. Spit tobacco leukoplakias tend to regress or to disappear with cessation. Cessation is difficult to achieve with an established habit because a "typical" pinch of spit tobacco contains approximately the same amount of nicotine as three cigarettes.
The risk of carcinoma or premalignant change in spit tobacco leukoplakias is not known. However, tobacco itself has been estimated to carry a four times greater risk of oral carcinoma development than that to be expected in populations not using tobacco.
A few dysplasias have been seen in biopsies of spit tobacco leukoplakias, but there are unquestionable cases of squamous carcinoma and verrucous carcinoma related to spit tobacco use (Figure 6). Therefore, in the assessment of a spit tobacco leukoplakia, biopsy should always be performed to establish a definitive diagnosis of the lesion at that point in time. If no dysplasia is found, then elimination of the habit and follow-up should suffice. If a patient cannot or will not give up the habit, then close follow-up and re-biopsy should be performed when progressive change such as thickening is seen or if evolution of a red component is found.
Leukoplakias of oral mucosa are common and readily recognized. Most leukoplakias represent benign keratosis, but these cannot be distinguished on clinical grounds from those that are premalignant. Only biopsy can make that determination reliably. Clinicians must develop criteria to use to determine which leukoplakias must be biopsied. Among the most important of these are anatomic site, patient age, risk factors, relationship to a probable irritant, and lesion response to removal of the suspected irritant. More frequent use of biopsies to assess leukoplakias will result in earlier diagnosis of oral cancer and, even more importantly, prevent oral cancer by virtue of intercepting lesions in their premalignant state.
Erythroplakia
The term erythroplakia is derived from erythroplasia. While erythroplasia has a diagnostic connotation in dermatology, erythroplakia as used in oral pathology is a clinical descriptor only. Erythroplakia may be defined as "a persistent, velvety red patch that cannot be identified as any other specific red lesion such as inflammatory erythemas or those produced by blood vessel anomalies or infection."9
Erythroplakias are less commonly seen than leukoplakias, but they are more ominous when identified. For example, Waldron and Shafer studied 58 cases and found that 51 percent were early invasive carcinoma and 41 percent were either carcinoma in situ or severe dysplasia.10 Similarly, Mashberg and colleagues described erythroplakia as the earliest sign of asymptomatic oral cancer.11
As mentioned previously, Waldron and Shafer found that of the 3,360 leukoplakias in their study, 670 were either dysplasia or carcinoma. The majority of those (413) represented at most mild or moderate degrees of dysplasia microscopically. Subsequently, the same authors studied erythroplakias and found that 92 percent were either early carcinoma or severe forms of dysplasia. Why the difference in severity between white and red lesions? The answer lies in the fact that leukoplakias are white principally due to hyperkeratosis. In the process of dysplasia, the epithelial cells tend to progressively lose differentiation, including their ability to keratinize. Also, as the cells become progressively more atypical, the body may mount an inflammatory reaction against the threat that the progress of the dysplasia represents, increasing the red component. Thus, erythroplakias tend to be composed of cells that are less mature, more atypical, and of a higher microscopic degree of dysplasia than leukoplakias. The loss of keratinization and increase in DNA content per unit area that accompanies increasing degrees of dysplasia and carcinoma is also the reason erythroplakias and carcinomas stain deeply with toluidine blue whereas leukoplakias do not (Figure 7).
The risk factors for erythroplakia are the same as those for oral carcinoma: tobacco use, alcohol abuse, and age older than 40 years. The oral sites where erythroplakias are most likely to occur are the oral floor, soft palate, tonsillar pillar, and lateral/ventral tongue (Figures 8 and 9).
There are relatively few considerations in the clinical differential diagnosis of erythroplakias. Burns will have an appropriate history and are usually painful. Inflammatory processes such as erythematous candidiasis will have an associated cause that can be removed, modified, or treated. In these situations, clinical improvement is to be expected within two weeks. Vascular abnormalities will generally be of long duration and tend to fade or blanch with pressure. Something virtually unique in the clinical appearance of an erythroplakia, but which may be difficult to appreciate without good light and a dry field, is a distinctly velvety surface texture (Figure 10).
High-risk patients for oral cancer should receive the most critical examination that looks for even tiny lesions that fit the criteria for erythroplakia. When observed, these should be biopsied without delay.
Speckled Leukoplakia
Speckled leukoplakia, also known as erythroleukoplakia, is a clinical hybrid lesion. Most authors describe it as a leukoplakia with interspersed red areas (Figure 11). Conversely, it may be an erythroplakia with a few white spots that do not wipe off (Figure 12). In truth, both descriptions are apt. Speckled leukoplakias are as dangerous as erythroplakias in terms of the tendency to represent more serious dysplasia or cancer on biopsy. It is of clinical relevance then that when selecting a site for an incisional biopsy in a leukoplakia, any red component should be included in the sample.
Actinic Keratosis
Squamous carcinoma of sun-exposed skin is a common disease and is second in incidence only to basal cell carcinoma. Actinic radiation is a potent carcinogen. The process by which carcinogenesis occurs in sun-exposed skin is comparable in most respects to that in oral mucosa. The cutaneous process includes a premalignant phase often termed actinic keratosis by dermatologists.
The vermilion border of the lips is a transition site between skin and mucosa but is somewhat more closely related to skin. Both the upper and lower vermilion borders are exposed to sunlight, but the lower vermilion receives substantially greater doses of actinic radiation. Therefore, it is the lower vermilion border that is at greatest risk for sun-induced cancer and for its premalignant counterpart, actinic keratosis.
People at greatest risk of developing labial actinic keratosis are older men with a fair complexion who spend considerable time out of doors. Women, presumably because of the protective effect of cosmetics, have less risk.
Along with the carcinogenic effect of ultraviolet radiation is a tendency for this radiation to cause damage to elastic fibers and to produce collagen degeneration. The result of the former is to induce vertical wrinkling or fissuring from the vermilion surface into adjacent skin. The result of the latter is to produce a thickened, puffy, and less pliable lip.
Typically, a person with labial actinic keratosis will present with one or more chronic crusting, keratotic patches on the lip, in addition to the above-described wrinkling and thickening (Figure 13). The patient will relate no symptoms but will often say the crusts can be peeled off but always return. Occasionally, a patient will have been told by a health care practitioner that the problem is "chronic herpes" and may have been prescribed an antiviral medication that has no beneficial effect. Typically, the crusts are not indurated. If induration is present, it is likely that carcinoma has developed.
The diagnosis of actinic keratosis can be made only by biopsy. The biopsy should include sufficient depth to allow the pathologist to be certain there is no invasion.
Treatment of labial actinic keratosis includes a minimum of complete excision of the lesion with a border of normal tissue and an admonition to protect the lip from sunlight, including use of a highly rated sunscreen. If the process is multicentric or if recurrence or a new lesion develops, then consideration should be given to vermilionectomy. The latter, if carried out as a plastic procedure to reduce the thickness of the sun-damaged collagen of the lip, results in a thinner lip whose new vermilion surface is composed of skin. The end result is usually very cosmetically and functionally acceptable (Figures 14 and 15).
Conclusion
The concept of oral premalignancy is well-established and clearly occupies the middle ground in the evolution of cancer from normal tissue. The process, as it involves the oral mucosa, is little different from that occurring in the bronchial tree, the uterine cervix, or the colon. What is significantly different is that the oral mucosa can be readily examined without special techniques or inconvenience to the patient.
The dentist is most familiar with the normal appearance of the oral mucosa, is best equipped to examine it, and is the individual with greatest opportunity to intercept premalignant oral disease.
Biopsy is the only method available to the profession that can unequivocally establish a definitive diagnosis of a clinical abnormality. When an abnormal lesion is found in a patient at risk for oral cancer, when an abnormality fails to respond to removal of a presumed local irritant, or when no local factor is present to account for a lesion, the dentist should not hesitate to perform a biopsy or to promptly refer the patient to a dental specialist for biopsy.
It must be realized that the dentist who detects, diagnoses, and causes a premalignant lesion to be effectively treated has essentially prevented his or her patient from developing oral cancer and has likely contributed significantly to saving that patient’s life.
Author
Raymond J. Melrose, DDS, is a professor in and the chairman of the Department of Oral and Maxillofacial Pathology at the University of Southern California School of Dentistry. He is also president-elect of the California Division of the American Cancer Society.
References
1. Paget J, Cancer following ichthyosis of the tongue. Trans Clin Soc London 3:88, 1870
2. Bouquot JE, The pathology and progression of oral pre-malignancy. Epithelial dysplasia symposium. 5th International Congress on Oral Cancer London, Great Britain 1977, p 1.
3. Bouquot JE, Gnepp DR, Laryngeal pre-cancer -- A review of the literature, commentary and comparison with oral leukoplakia. Head Neck 13:488-97, 1991.
4. Bouquot JE, Whitaker SB, Oral leukoplakia-Rationale for diagnosis and prognosis of its clinical sub-types or "phases." Quintessence Int 25:133-40, 1994.
5. Waldron CA , Shafer WG, Leukoplakia re-visited: A clinicopathologic study of 3256 oral leukoplakias. Cancer 36:1386-92, 1975.
6. Greenspan D, Greenspan JS, Significance of oral hairy leukoplakia. Oral Surg Oral Med Oral Pathol 73:151-4, 1992.
7. Damm D, Curran A, White D, Leukoplakia of the maxillary vestibule -- An association with Viadent? Oral Surg Oral Med Oral Pathol Oral Radiol Endod 87:61-6, 1999.
8. Hansen LS, Olson JA, Silverman S Jr, Proliferative verrucous leukoplakia -- A long-term study of thirty patients. Oral Surg Oral Med Oral Pathol 60:285-98, 1985.
9. Wood NK, Goaz PW, in Differential Diagnosis of Oral and Maxillofacial Lesions, 5th ed. C V Mosby, St Louis, 1997, p 57.
10. Waldron CA, Shafer WG. Erythroplakia of the oral cavity. Cancer 36:1021-8, 1975.
11. Mashberg A, Morrissey JB, Garfinkel L, A study of the appearance of early asymptomatic oral squamous carcinoma. Cancer 32:1436-45, 1973.
To request a printed copy of this article, please contact/Raymond J. Melrose, DDS, Oral Pathology Associates, Inc., 11500 W. Olympic Blvd., Suite 390, Los Angeles, CA 90064.
Legends
Figure 1. A thin, homogenous leukoplakia of the lateral tongue in a 42-year-old male cigarette smoker. A biopsy showed moderate epithelial dysplasia. Criteria other than clinical appearance must be used to assess such lesions.
Figure 2. A thick leukoplakia. A biopsy showed moderate epithelial dysplasia.
Figure 3. A rough leukoplakia of the maxillary gingiva. There were no risk factors and no discernable local irritants. A biopsy showed benign verrucous keratosis, but the lesion continued to enlarge. Several years later, another biopsy revealed moderate epithelial dysplasia confirming that that this is a case of proliferative verrucous leukoplakia.
Figure 4. Extensive proliferative verrucous leukoplakia of the mandibular gingiva. The process had been evolving for years, and ultimately the patient developed verrucous carcinoma in the area.
Figure 5. Wrinkled and fissured leukoplakia lines a "pocket" into which spit tobacco is packed.
Figure 6. Squamous carcinoma of the maxillary alveolar mucosa and vestibule has developed in this area where spit tobacco has been placed for years.
Figure 7. Positive toluidine blue staining of an extensive ventral tongue lesion in a patient successfully treated for squamous carcinoma of the soft palate 12 months prior. The patient continues to smoke. He was referred for a new denture by his radiation oncologist.
Figure 8. Erythroplakia of the oral floor in a patient at high risk for squamous carcinoma. A biopsy revealed severe epithelial dysplasia.
Figure 9. Erythroplakia of the soft palate in a patient at high risk for squamous carcinoma. A biopsy of this case also revealed severe epithelial dysplasia.
Figure 10. This erythroplakia of the lateral tongue shows a distinct margin between normal smooth texture and the velvety appearance often seen in erythroplakia. A biopsy showed carcinoma in situ, the most severe degree of dysplasia.
Figure 11. Speckled leukoplakia may seem to be leukoplakia with a red component. A biopsy of this lesion revealed severe epithelial dysplasia.
Figure 12. Speckled leukoplakia may also seem to be a red lesion with white foci. A biopsy of this lesion also revealed severe epithelial dysplasia.
Figure 13. Actinic keratosis with dysplasia was found in a biopsy of the crusted area. Note the thick, puffy contour, vertical fissuring and loss of definition of the skin/vermilion junction.
Figure 14. Pre-operative photo of an older man with biopsy-proved actinic epithelial dysplasia.
Figure 15. Postoperative photo after treatment by vermilionectomy. Note the reduction of contour and excellent cosmetic result.
