Erythroplakia is the clinical diagnostic term A chronic red mucosal macule which cannot be given another specific diagnostic name and cannot be attributed to traumatic, vascular or inflammatory causes, i.e. it is a diagnosis of exclusion. Once a biopsy is performed the term may still be used but should be modified to included the presence or absence of atypical or dysplastic epithelial cells. The lesion is a precancer, i.e. it carries a higher than normal risk of malignant transformation.
History & More Definitions.
The twentieth century has seen the emergence and acceptance of the idea that certain lesions which are apparently benign and innocuous in appearance are in reality precancerous lesions and carry a greater than normal risk of transforming into, or in some other fashion "becoming," an invasive malignancy. Several such precancers are associated with the membranes of the head and neck (H&N) region, most notably actinic cheilosis, oral leukoplakia, laryngeal keratosis and erythroplakia (Table 1). Of these, erythroplakia is the least known to health professionals, even though it carries one of the highest cancer risks and has been discussed in print for more than eighty years. Much has been learned about erythroplakia and its typical microscopic counterpart, carcinoma in situ, during the past several decades, enough to now be able to present a rather complete picture of the clinical, pathologic and epidemiologic characteristics. Before the century is over, then, it seems appropriate to review our knowledge and understanding of this dangerous and intriguing premalignancy.
A review of premalignant lesions should begin with a number of definitions because considerable confusion has arisen as our understanding of precancers has evolved. Fortunately, the World Health Organization (WHO) has provided simple but workable definitions of oral precancerous conditions and lesions. The following variations of the WHO definitions, from the U.S. textbook standard by Neville et al, are recommended for use with oral precancers and will be used throughout this paper:
Precancerous lesion (precancer, premalignancy, potentially malignant lesion) - An apparently benign, morphologically altered tissue which has a greater than normal risk of containing a microscopic focus of cancer at biopsy or of transforming into a malignancy after diagnosis.
Precancerous condition - A disease or patient habit which does not necessarily alter the clinical appearance of local tissue but is known to have a greater than normal risk of precancer or cancer development.
Malignant transformation rate/potential - The risk of cancer being present in a precancerous lesion or condition, either at the time of initial diagnosis or at a future date. The potential for mucosa without precancerous lesions or conditions is termed "normal."
Leukoplakia - A chronic white mucosal macule which cannot be scraped off, cannot be given another specific diagnostic name, and does not disappear with removal of potential etiologic factors (excepting tobacco).
Erythroplakia - A chronic red mucosal macule which cannot be given another specific diagnostic name and cannot be attributed to traumatic, vascular or inflammatory causes.
Carcinoma in situ (CIS, intraepithelial carcinoma) - An histopathologic diagnosis defined as a proliferation of primitive basal epithelial cells from the basement membrane to the surface, with almost all cells manifesting cytologic atypia generally accepted as characteristics of malignancy. Immediate maturation into a thin keratin layer is possible, but no invasion into underlying connective tissues can be seen.
Severe epithelial dysplasia - A histopathologic diagnosis defined as a proliferation of basal cells into but not completely through the upper third of the epithelium, with most cells manifesting cytologic atypia generally accepted as characteristics of malignancy. No invasion into underlying tissues can be seen.
Superficially invasive squamous cell carcinoma - A histopathologic diagnosis of a routine squamous cell carcinoma, almost always well differentiated, which has invaded only slightly (5 mm. or less) into underlying connective tissues.
It is especially important to remember that a premalignancy is not guaranteed to transform into cancer, as was once the common belief. Many, in fact, do so only in a small proportion of cases, forcing the clinician to make some very real choices relative to the management of such lesions.
A diagnosis of exclusion. It should be emphasized at the outset that erythroplakia, like leukoplakia, is a diagnosis of exclusion which requires the clinician to rule out all other erythematous oral lesions before the term can be applied to a specific lesion in a particular patient. Such entities as mechanical trauma, chemical or thermal injury, infectious/allergic/autoimmune mucositis (plasma cell gingivitis, erythema multiforme, psoriasis, etc.), submucosal hemorrhage or vascular anomaly, migratory glossitis and migratory stomatitis, median rhomboid glossitis, denture sore mouth, atrophic glossitis (vitamin deficiency, xerostomia, atrophic candidiasis, etc.), Wegener's granulomatosis (gingival location), and macular mucosal Kaposi's sarcoma must all be considered in the differential diagnosis. In the larynx, inflammatory erythema from smoking ("smoker's larynx") and/or vocal cord trauma must also be excluded.
It should additionally be emphasized that the term erythroplakia is a clinical one, even though biopsy evaluation almost always is able to identify dysplastic or malignant cells at the microscopic level. The presence of such cells does not alter the clinical diagnosis and erythroplakia is still erythroplakia with or without dysplastic epithelial cells, so long as it cannot be classified as a different form of red mucosal lesion.
The initial description of erythroplakia was a penile lesion reported in 1911 by Queyrat under the term "erythroplasia," although Fournier and Darier may have first described a penile case in 1893 and Gibb may have described a similar lesion of the vocal cords as early as 1865. The first microscopic analysis of carcinoma in situ was reported in 1910 by Rubin as a distinctive "incipient carcinoma" of the uterine cervix. During the next decade Darier et al published the first H&N example of "erythroplasia of Queyrat," a macular lesion of the buccal mucosa. In 1932 Broders coined the term "carcinoma in situ," using a laryngeal example, and emphasized its preinvasive status just a few years after Schiller had crystallized the theory of preinvasiveness for the uterine cervix. In contrast to today's understanding of the disease, these early authors presumed that CIS would eventually invade and metastasize in every cases, i.e. it was already a malignancy rather than a premalignancy.
The first emphasis on the high cancer risk of mucosal erythema or "inflammation" in vocal cord leukoplakia (laryngeal keratosis) was published in 1942 by Graham and the early literature on head and neck CIS was considerably dominated by laryngeal cases. Gorlin, in his classic 1950 paper reported the first collection of cases of oral CIS in order to counter the prevalent belief that the larynx was essentially the only H&N site of involvement. Shedd et al joined him with an additional series of cases and the flurry of clinical research which followed allowed Shafer and Waldron in the 1970s to provide an extensive clinicopathologic characterization of oral lesions and to recommend the use of the term "erythroplakia" as the most appropriate clinical diagnosis for the red mucosal macules. Miller, Bauer and Kambic provided similar influential reviews for laryngeal CIS.
Recent emphasis has also been placed on the red mucosal changes associated with oral erythroleukoplakia, the leukoplakia type or phase with the highest malignant transformation rate. The epidemiologic characteristics of CIS were delineated in the late 1980s, confirming that it is, indeed, a rare entity. As few as 1,300 H&N cases are diagnosed annually throughout the United States.
Much insightful information relating to premalignancy has been gleaned from the study of H&N erythroplakia and CIS, despite the fact that clinicopathologic investigations have included very few cases. These lesions have frequently been found adjacent to invasive carcinoma of the mouth and other H&N sites, and Mashberg and Samit have provided an invaluable service by stressing the innocuous erythematous nature of early oral cancers and their adjacent tissues. The juxtapositioning of these preinvasive and invasive lesions of the mouth was also an important impetus for the early development of the well-known "field cancerization" theory of Slaughter et al.
Frequent occurrence in close proximity to carcinoma, of course, only suggests a malignant transformation role for erythroplakia. Follow-up studies of a number of these precursor lesions must demonstrate a greater than normal risk of transformation or "degeneration" into an invasive malignancy. No prospective studies of purely red mucosal lesions has been published. A few investigations, however, have followed a total of 272 patients with microscopically proven severe epithelial dysplasia and/or carcinoma in situ of the mouth, beginning with the Tennessee study by Mincer et al. in 1972 (Table 2). These studies have determined a malignant transformation rate of 26%, but the type and extent of treatment is unclear for most reported cases and so there is still some doubt as to the validity of the rates. Follow-up studies of laryngeal CIS have been in the literature since 1952 and have found a similar transformation rate for treated lesions at that site (Table 3).
Oral and vermilion cases. The causes of erythroplakia are assumed to be the same as the causes of invasive squamous cell carcinoma. More than 94% of CIS patients, for example, smoke tobacco and more than 65% abuse alcohol. On the lower lip, ultraviolet irradiation damage, called actinic cheilosis or actinic cheilitis, plays an important etiologic role for CIS. Almost all vermilion cases have microscopic evidence of solar elastosis of the submucosal connective tissues.
Candida albicans has often been demonstrated in erythroleukoplakia and the red component (the white component as well) has been shown to diminish or disappear after antifungal therapy in some cases. The clinical erythema in such cases may be the result of inflammation rather than dysplasia, although Barrett et al have shown a positive correlation between the severity of dysplastic epithelial cells and the presence of the fungus.
Human papillomavirus (HPV) is a second micro-organism implicated in mucosal carcinoma of the H&N region, but the significance of this association remains unclear. The virus is certainly a common feature of CIS of the uterine cervix, and HPV-16 and HPV-18 infection of in vitro cervical epithelium (otherwise sterile) has been shown to produce quite dysplastic cells. The virus has also been demonstrated in as many as 40% of oral leukoplakia lesions. However, HPV is also found in a certain proportion of normal oral mucosa and recent investigations have determined a much smaller proportion of HPV-affected tissue in leukoplakia cases. To date, no positive correlation has been specifically reported between HPV and CIS or severe epithelial dysplasia of the H&N, although a rare and unique papillary form of oral CIS, bowenoid papulosis, seems to be obviously associated.
The Epstein-Barr virus (EBV) has also been reported in 50% of oral erythroplakias and 40% of oral CIS lesions. Again, the significance of this in unclear and it is not known whether the EBV is acting as a carcinogen, simply exists (thrives?) within altered or otherwise damaged lesional cells, or is being shed at higher levels because of tumor-related immunodepression.
Tumor-suppressor genes and other cell maturation factors are involved in the initiation and progression of a wide variety of cancers, including H&N carcinoma. These have not been specifically investigated relative to erythroplakia or CIS, but Table 4 lists those with the strongest reported association with dysplastic H&N epithelium. To date, these factors have contributed little of practical application to precancers of this region but it is expected that future investigations will confirm prognostic correlations of importance.
Laryngeal cases. Virchow was probably the first to suggest that voice and alcohol abuse could produce precancers of the vocal cords, but since his time surprisingly few investigators have addressed etiology in their research. Epidemiologic evidence has, however, strongly implicated tobacco smoking in the etiology of laryngeal carcinoma and autopsy studies have demonstrated a strong correlation between daily tobacco consumption and the presence of laryngeal dysplasia. Almost half of all persons smoking more than 20 cigarettes daily have dysplastic vocal cord mucosa at autopsy, compared to 4% for nonsmokers. Unfortunately, no study has specifically addressed erythroplakia or CIS.
Tobacco chewing also appears to be a significant risk factor for laryngeal carcinoma, but only in persons who chew all day, every day. Alcohol abuse has been shown to have a synergistic association with tobacco abuse in the production of laryngeal carcinoma, but no data is available relative to laryngeal precancers. Available data somewhat implicate HPV in the etiology of some laryngeal carcinomas, but this has not been addressed for CIS.
CIS of all anatomic sites annually affects 26.1 persons per 100,000 population (4.5 for males and 46.9 for females), excluding skin cases. Ninety percent of these are found at only four body sites: uterine cervix, breast, colon and corpus uteri. H&N lesions account for only 2% of the total, although the H&N is unique in being the only site with a strong male predilection.
Oral cases. There are no annual incidence data (number of new cases diagnosed annually) for oral erythroplakia, but the few epidemiologic investigations which have determined point prevalence rates (number of cases in a population at a given point in time) for nondescript red mucosal patches have found them in 1-17 per 1,000 adults. One of these studies distinguished between inflammatory and precancerous red patches, finding the latter to represent half of focal mucosal erythemas in a cohort of dental school patients (erythroplakia prevalence: 3 per 1,000 males, 1 per 1,000 females, 2 per 1,000 both genders). The erythroplakia prevalence rate of 17 per 1,000 U.S. adult tobacco users established by Prout et al also excludes an additional 43 per 1,000 inflammatory/traumatic red patches.
CIS, including severe epithelial dysplasia for purposes of discussion in this section, represents 80-90% of oral true erythroplakias and several H&N incidence investigations are available for CIS/severe dysplasia. These studies have shown that biopsy-proven oral CIS is a rare disease, diagnosed in only 0.1-0.6 of every 100,000 persons annually in the U.S., 0.2-1.0 if oropharyngeal lesions are included.
CIS usually accounts for only 2% of oral carcinomas but in communities with a strong health consciousness and early detection procedures in place, this proportion could reach as high as 17%. With the contemporary emphasis on early cancer detection and routine oral soft tissue examinations, the relative frequency of oral cancers found in this premalignant/preinvasive stage is greater today than it was half a century ago, and the incidence of CIS has increased accordingly, even as the incidence of invasive oral cancers has slightly decreased.
Men are affected with intraoral CIS four times more frequently than women (annual incidence rates: 0.4-1.0 vs 0.1-0.2 per 100,000 males and females, respectively).30-32 This is drastically different from the gender predilection of lesions of the urogenital tract where females are much more frequently affected, largely because the uterine cervix is such a common site for this disease.30 Persons 55-75 years of age are most often affected by oral CIS and this disease is almost unheard of before the fourth decade of life (Figure 1A). In contrast, CIS of the uterine cervix typically affects women in the 25-44 year age bracket and more than 10% of cases occur in 15-24 year olds.31
There seem to be only minor differences between racial and ethnic groups relative to oral CIS, with Hispanics, American Indians and Japanese Americans showing the lowest incidence among U.S. population subgroups.31 There also seems to be a slight urban/rural variation in frequency, with rural populations demonstrating somewhat lower incidence rates than urban populations.
Vermilion cases. Most cases of CIS of the vermilion borders of the lips are not included in federal epidemiology studies, but where effort has been made to identify such cases it has been shown to be the most common site of H&N involvement, accounting for 42% of all cases.32 Vermilion lesions are diagnosed in 1.0 of every 100,000 persons annually in the U.S. Approximately 88% of vermilion CIS occurs in males and the average age at diagnosis is 70 years, almost a decade older than the average age of persons with oral lesions (Table 5). Incidence at this site continues to increase throughout life (Figure 1B).
Laryngeal cases. There are no statistics relative to the frequency of laryngeal erythroplakia in any population. Annual incidence rates, however, have been published for CIS of this site, showing a rate that is less than the rate for oral CIS: 0.2-0.4 cases per 100,000 persons each year. The laryngeal lesions are much more common in males (88%) than in females and are usually diagnosed in persons 60-80 years of age (Figure 1C), somewhat older than the age at diagnosis for the oral counterpart.
Erythroplakia, as the name implies, is an asymptomatic red macule or patch on a mucosal surface. The reason for the red color is unclear, but seems to be related to a combination of dilation and engorgement of the subepithelial microvascular system and a thinning of the keratin layer or of the entire epithelium. The vascular change is presumably part of an immunologic or inflammatory response to the dysplastic epithelial cells, and the lack of keratin seems to be a simple case of inadequate cellular maturity to produce this end product.
All mucosal surfaces of the upper aerodigestive tract are susceptible to erythroplakia and CIS development. Almost half of all CIS cases, however, occur on vermilion or intraoral surfaces, with the rest being evenly divided between the larynx and the pharynx. The most frequent oral sites of involvement, in decreasing order of frequency, are the lips (vermilion), lateral and ventral tongue, oral floor and soft palate. In the larynx, almost all cases occur on the vocal cords. The least likely site of H&N involvement is the nasopharynx and, in fact, not a single case of erythroplakia or CIS has thus far been reported from that location.
Oral cases. The typical oral lesion is less than 1.5 cm. in greatest diameter, but examples larger than 4 cm. have been seen. It may be quite sharply demarcated from the surrounding pink mucosa or it may gradually blend into normal tissues (Figure Figures 2A, 2B). Its surface is typically smooth and regular in coloration, but occasional lesions demonstrate a pebbled or granular surface change similar to that seen in granular or phase III leukoplakia. Rarely, a true papilloma-like appearance is seen (papillary CIS, bowenoid papulosis) with multiple, clustered erythematous patches, although the papillary nature may be clinically indistinct (Figure 2C).
Smooth erythroplakia is soft to palpation, often described as having a velvety feel to the examiner with hands unencumbered by latex gloves. The pebbled lesions tend to be somewhat firm, but erythroplakia never actually becomes hard or truly indurated until a deeply invasive carcinoma develops within it.38 Superficially invasive squamous cell carcinoma is likewise seldom palpable.
Erythroplakia may be admixed with or adjacent to leukoplakia in the mouth, in which case it is usually designated as erythroleukoplakia and considered to be part of the spectrum of leukoplakic lesions or phases (Figure 2D). The red areas of erythroleukoplakic lesions are the sites most likely to contain or to develop dysplastic cells and should, therefore, be the sites most readily biopsied and most carefully examined clinically. The clinical localization of dysplastic epithelial cells with enlarged, hyperchromatic nuclei is aided by coating suspicious mucosa with toluidine blue. After rinsing with acetic acid, residual stain indicates areas with immature cells (Figure 3). The clinician must be cautious, however, because many false positive toluidine tests are produced by inflammatory erythema.
The clinician must also be thorough in his or her clinical examination of adjoining mucosa because erythroplakia and erythroleukoplakia are seen adjacent to or part of 33% and 60%, respectively, of all early squamous cell carcinomas of the mouth. While it has been shown that oral carcinomas with associated or adjacent leukoplakia behave in a less aggressive fashion than carcinomas without such lesions, no similar data are available relative to carcinomas associated with erythroplakia.
Vermilion cases. Almost all examples of vermilion CIS are associated with actinic cheilosis. Actinic irradiation damage frequently produces a vermilion border with a generalized but blotchy intermixing of pale and erythematous areas. The erythema, however, typically has a bluish hue and there is usually a surface keratin layer which prevents the appearance of the bright red change so characteristic of internal mucosal erythroplakia. More than half of all CIS lesions of the vermilion border, moreover, are ulcerated at the time of biopsy, probably secondary to the minor but chronic trauma of holding a cigarette in the area. Ulceration is a decidedly uncommon occurrence with the internal mucosal erythroplakic or CIS lesion unless it is associated with an obvious oral carcinoma (Table 5).
Laryngeal cases. Erythroplakia and erythroleukoplakia have been largely ignored in the laryngeal literature. This has occurred in part because red mucosa has long been considered to be simple inflammatory hyperemia, which is an almost constant feature of "smoker's" larynx. Ironically, a few authors have actually included mucosal erythema as a routine and required part of their definition of a white lesion, laryngeal keratosis or leukoplakia. Beginning at the century's midpoint, however, at least a few authors warned that the more heavily "inflamed" keratoses were at greatest risk of dysplastic or malignant change. Now, of course, the redness associated with laryngeal keratosis is well accepted as a dysplastic rather than purely inflammatory feature and is seen in 16% of keratoses, according to population studies. It seems to be unusual for a purely red vocal cord lesion to be biopsied prior to invasion, those with this appearance account for less than 15% of the CIS lesions of the larynx.
Almost all erythroplakic lesions contain dysplastic cells. The histopathology may be only mild or moderate epithelial dysplasia but more typically it is CIS or severe epithelial dysplasia. A certain proportion also demonstrate microinvasive squamous cell carcinoma. This is usually seen in only 5-10% of cases, but in one clinicopathologic review of 58 cases more than half were invasive.
Microscopic features of CIS are the same for oral, pharyngeal and laryngeal lesions. The two essential criteria for a diagnosis of this disease are (Figure 4):
Nuclei are typically hyperchromatic and enlarged, with the amount of cytoplasm diminished. Lesional cells are usually quite similar one to another (Figures 4A, 4B), but extreme variation in cellular size and shape may be present in some areas (Figure 4C). Mitotic activity is sometimes pronounced and abnormal mitotic figures may be noted. Keratin pearls are almost never seen.
The purists among us do not accept even the smallest hint of surface keratinization in their diagnosis of CIS, but most pathologists appear to accept the presence of flattened keratinocytes on the surface, with abrupt transition from the underlying dysplastic cells, as part of the disease (Figure 5). Vermilion CIS almost always demonstrates a thin layer of surface keratin, and in this regard is quite similar to Bowen's disease or CIS of sun-damaged skin.
The lateral transition from normal epithelium to CIS is often abrupt, but the altered epithelium may gradually "shade off" into the adjacent normal epithelium without a clear zone of demarcation (Figure 6). Additional areas of dysplasia may be seen at remote sites within the epithelium. The pathologist should carefully evaluate all such areas for signs of invasion because multifocal carcinoma is not a rare event in the H&N region.
As with much dysplastic epithelium, rete pegs frequently are bulbous or tear-drop shaped, often with secondary proliferations or projections of abnormal cells (Figures 4B, 6A, 7). These abnormal shapes are considered to be signs of dysplasia whether or not atypia is found in the individual cells of the rete ridges. The basement membrane must, moreover, be carefully examined for areas of micro invasion. This evaluation can be quite problematic in lesions associated with strong inflammatory (immunologic?) responses, which may result in a degenerated basement membrane. This especially a problem in the rare entity called "lichenoid dysplasia." All the usual features of epithelial dysplasia are present in this disease but the lowest layers of the epithelium demonstrate liquefactive or ballooning degeneration and saw-toothed or completely missing rete ridges. There is a characteristic subepithelial band of chronic inflammatory cells, predominantly T-cells. It is not yet known whether this variant represents premalignant change in a true lichen planus or is simply an unusual immune response to the dysplastic cells.
An occasional interpretive difficulty in the lower regions of CIS-involved epithelium is extension or "creep" of dysplastic epithelium down the ducts of underlying minor mucous glands (Figure 8). While relatively easy to identify, the significance of this feature is not yet clear. Some authorities consider it to be a sign of early invasion while others see it as simple in situ extension of lesion cells in a fashion not unlike the lateral extension at the margins of the lesion.3
Severe epithelial dysplasia may blend subtly into CIS, and some lesions are extremely difficult to differentiate (Figure 7). This difficulty has prompted some authors to suggest combining both microscopic diagnoses into a single entity, usually referred to as grade III dysplasia or grade III oral intraepithelial neoplasia (OIN III). Such a grading system maintains the two less severe types of epithelial dysplasia but renames them as grade I (mild dysplasia) and grade II (moderate dysplasia) intraepithelial neoplasia. Although a classification system with some merit, this grading of a continuum of neoplastic development appears not to be popularly accepted.
In routine epithelial dysplasia, i.e. those cases not thought to be CIS, there is a gradual transition from abnormal cells of the lower epithelium into normal or more mature superficial epithelial cells. When the atypical cells remain within the lowest third of the epithelium, a diagnosis of mild dysplasia is used. When such cells extend from the basement membrane to the middle third of the epithelium, a diagnosis of moderate dysplasia is used.
Only when primitive or atypical cells extend above the middle third of the epithelium is the diagnostic term severe (premalignant) epithelial dysplasia used. This implies that dysplastic change does not extend completely to the surface of the epithelium but, in truth, many lesions show atypical cells so close to the surface that it is impossible to accurately distinguish between severe dysplasia and CIS. The distinction, fortunately, appears to be moot, as both lesions have a similar biological behavior.
Oral and vermilion cases. Erythroplakia should always be removed or destroyed, either as an excisional biopsy or after an incisional biopsy has demonstrated dysplastic cells. A conservative surgical procedure such as mucosal stripping is usually performed, with minimal damage to deeper connective tissues. This has the distinct advantage of preserving tissue for microscopic evaluation of potential regions of invasion, but destructive techniques such as laser ablation, electrocoagulation and cryotherapy have proven to be effective. Some surgeons advocate a more aggressive approach on the assumption that there may already be an invasive carcinoma present.
The key to therapy in this disease is extended clinical follow-up and patients should be examined every 3 months for the first year and semi-annually for an additional 4 years. Follow-up investigations of hundreds of lesions have demonstrated a malignant transformation rate of 14-50% (Table 2). This rather broad range may be due to the use of different histopathologic criteria, different biases in case selection, or real differences in various patient groups. Results must also be strongly influenced by the fact that most lesions are completely removed once a microscopic diagnosis is made. Recently, Lumerman et al47 compiled a table of 91 patients with oral epithelial dysplasia (all types) who received no further surgical treatment after the initial biopsy. Their results indicated that 58% remain the same, 9% increase in size and 15% develop invasive carcinoma.
Laryngeal cases. CIS of the larynx has been more extensively followed than CIS of the mouth or pharynx. Overall, 29% of treated laryngeal lesions eventuate in invasive disease, while untreated cases transform at a much higher rate (Table 3). Laryngeal lesions should all be treated and subjected to the same follow-up schedule as oral lesions.
Is It Cancer or Is It Not?
We believe that it is best to consider CIS as a premalignancy rather than true carcinoma, as not all lesions will become invasive malignancy if left alone (Table 3). Its biological behavior, moreover, seems to be strongly influenced by its location. Santa Cruz has emphasized, for example, that Bowen's disease of the skin has the exact histopathology as a CIS of the mucous membranes but it seldom becomes invasive. This observation dramatically highlights another of his conclusions that "few topics in pathology are as controversial as carcinoma-in-situ." The situation is greatly compounded for H&N cases by the fact that there is poor intraexaminer and interexaminer reliability or consistency in the diagnosis of epithelial dysplasia.
There is no information relative to the length of time an oral CIS typically exists before it begins to invade into underlying connective tissues, or the proportion which will actually invade if left untreated. Three laryngeal CIS studies, however, have followed untreated lesions and have determined a malignant transformation rate ranging from as low as 33% to as high as 90% (Table 3). In other words, no investigation has yet proven that CIS of any H&N site is guaranteed to invade underlying tissues or metastasize to distant anatomic regions. Even in the uterine cervix, where the vast majority of CIS cases occur, various follow-up studies have shown a wide range of malignant (invasive) transformation rates, from 21% to 90%.
Under the microscope, erythroplakia typically presents as CIS, severe epithelial dysplasia, or occasionally as superficially invasive carcinoma. The relative frequency of each of these diagnoses is variable, depending on whether one looks at an entire population, a dental school's patient pool, or a high risk group such as patients in a Veterans Administration hospital. In high risk groups, 90% of erythroplakias are CIS, 7% are severe epithelial dysplasia and 3% are superficially invasive squamous cell carcinomas.
References (Listed by Authors)
Abbey LM, Kaugars GE, Gunsolley JC, et al. Intraexaminer and interexaminer reliability in the diagnosis of oral epithelial dysplasia. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995; 80:188-191.
Altman F, Ginsberg I, Stout AP. Intraepithelial carcinoma (carcinoma in situ) of the larynx. Arch Otolaryngol 1952; 56:121-133.
Amagasa T, Yakoo E, Sato K, et al. A study of the clinical characteristics and treatment of oral carcinoma in situ. Oral Surg Oral Pathol Oral Pathol 1985; 60:50-55.
Axell T, Pindborg JJ, Smith CJ, et al. Oral white lesions with special reference to precancerous and tobacco-related lesions: conclusions of an international symposium held in Uppsala, Sweden, May 18-21, 1994. International Collaborative Group on Oral White Lesions. J Oral Pathol Med 1996; 25:49-54.
Banoczy J, Csiba A. Occurrence of epithelial dysplasia in oral leukoplakia: analysis and follow-up study of 120 cases. Oral Surg Oral Med Oral Pathol 1976; 42:766-774.
Barrett AW, Kingsmill VJ, Speight PM. The frequency of fungal infection in biopsies of oral mucosal lesions. Oral Dis 1998; 4:26-31.
Bauer WC. Concomitant carcinoma in situ and invasive carcinoma of the larynx. Can J Otolaryngol 1974; 3:533-542.
Birchall M, Winterford C, Tripconi L, et al. Apoptosis and mitosis in oral and oropharyngeal epithelia: evidence for a topographical switch in premalignant lesions. Cell Prolif 1996; 29:447-456.
Bouquot JE. Common oral lesions found during a mass screening examination. J Amer Dent Assoc 1986; 112:50-57.
Bouquot JE. Epidemiology. In: Gnepp DG (ed). Pathology of the head and neck. New York: Churchill Livingstone, 1987: 263-314.
Bouquot JE. Reviewing oral leukoplakia-clinical concepts for the 1990s. J Amer Dent Assoc 1991; 122:80-82.
Bouquot JE. Leukoplakia and erythroplakia: a review and update. Pract Perio Aesth Dent 1994; 6:1-10.
Bouquot JE, Ephros H. Erythroplakia -- the dangerous red mucosa. Pract Perio Aesth Dent 1995; 7:59-68.
Bouquot JE, Gnepp D. Epidemiology of carcinoma in situ of the upper aerodigestive tract. Cancer 1988; 61: 1685-1690.
Bouquot JE, Gnepp DR. Laryngeal precancer--a review of the literature, commentary and comparison with oral leukoplakia. Head Neck 1991; 13: 488-497.
Bouquot JE, Kurland LT, Weiland LH. Carcinoma in situ of the upper aerodigestive tract: incidence, time trends and follow-up in Rochester, Minnesota, 1935-1984. Cancer 1988; 61:1691-1698.
Bouquot JE, Young S. Low PCR detection rate of human papillomavirus in 500 oral leukoplakia cases. 1994:unpublished data.
Bouquot JE, Weiland LH, Kurland LT. Leukoplakia and carcinoma in situ synchronously associated with invasive oral/pharyngeal carcinoma in Rochester, Minnesota, 1935-1984. Oral Surg Oral Med Oral Pathol 1988; 65:199-207.
Bouquot JE, Weiland LE, Kurland LT. Multiple primary carcinomas of the upper aerodigestive tract (UAT); the complete experience of one U.S. community, 1935-1988. J Cancer Res Clin Oncol 1990; 116:358.
Bouquot JE, Weiland LH, Kurland LT. Laryngeal keratosis and carcinoma in the Rochester, Minnesota population, 1935-1984. Cancer Detect Prevent 1991; 15:83-91.
Bouquot JE, Whitaker SB. Oral leukoplakia--rationale for diagnosis and prognosis of its clinical subtypes or "phases." Quint Internat 1994; 25:133-140.
Broders AC. Carcinoma in situ contrasted with benign penetrating epithelium. J Am Med Assoc 1932; 99:1670-1674.
Coltrera MD, Zarbo RJ, Sakr WA, et al. Markers for dysplasia of the upper aerodigestive tract. Suprabasal expression of PCNA, p53, and CK19 in alcohol-fixed, embedded tissue. Am J Pathol 1992; 141:817-825.
Crissman JD, et al. Preinvasive lesions of the upper aerodigestive tract: histologic definitions and clinical implications (a symposium). Pathol Ann 1987; 22:311-352.
Cruz I, van den Brule AJ, Steenbergen RD, et al. Prevalence of Epstein-Barr virus in oral squamous cell carcinomas, premalignant lesions and normal mucosa - a study using the polymerase chain reaction. Oral Oncol 1997; 33:182-188.
Darier J, Lemaitre F, Monier L. Les modes de debut des cancers de la bouche et des machiores. Bull Cancer (Paris) 1924; 13:256-272.
Dhingra JK, Zhang X, McMillan K, et al. Diagnosis of head and neck precancerous lesions in an animal model using fluorescence spectroscopy. Laryngoscope 1998; 108:471-475.
Diaz-Guzman L, Castellanos JL. Lesiones de la mucosa bucal. Estudio epidemiologico en 7,297 pacientes. Revista adm 1991; 48:75-80.
Doyle PJ, Flores A, Douglas GS. Carcinoma in situ of the larynx. Laryngoscope 1977; 87:310-316.
Elder D, Elenitsas R, Jaworsky C, Johnson B Jr. Lever's Histopathology of the skin, 8th edition. Philadelphia; Lippincott-Raven, 1997.
Elman AJ, Goodman M, Wang CC, et al. In situ carcinoma in the vocal cords. Cancer 1979; 43:2422-2428.
Emilion G, Langdon JD, Speight P, et al. Frequent gene deletions in potentially malignant oral lesions. Br J Cancer 1996; 73:809-813.
Eversole LR, Sapp JP. c-myc oncoprotein expression in oral precancerous and early cancerous lesions. Eur J Cancer B Oral Oncol 1993; 29B:131-135.
Field EA, Field JK, Martin MV. Does Candida have a role in oral epithelial neoplasia? J Med Vet Mycol 1989; 27:277-294.
Fournier A, Darier J. Epitheliome benin syphiloide de la verge. Bull Soc franc dermat et syph 1893; 4:324-332.
Fu YS, Reagan JW. Pathology of the uterine cervix, vagina, and vulva. Philadelphia: W. B. Saunders, 1989:235-258.
Gabriel CE, Jones DG. Hyperkeratosis of the larynx. J Laryngol Otol 1973; 87:129-134.
Gibb PG. Primary cancer of the larynx, destroying the epiglottis, as seen with the laryngoscope. Trans Path Soc London 1865; 16:54-57.
Girod SC, Krueger G, Pape HD. P53 and Ki67 expression in preneoplastic and neoplastic lesions of the oral mucosa. Int J Oral Maxillofac Surg 1993; 22:285-288.
Gorlin RJ. Bowen's disease of mucous membrane of the mouth; review of literature and presentation of cases. Oral Surg Oral Med Oral Pathol 1950; 50; 3:35-51.
Graham HB. Keratosis of the larynx. Arch Otolaryngol 1942; 36:735-739.
Gupta PC, Mehta FS, Daftary DK, et al. Incidence rates of oral cancer and natural history of oral precancerous lesions in a 10-year follow-up study of Indian villagers. Community Dent Oral Epidemiol 1980; 8:287-333.
Hellquist H, Lundgren J, Olofsson J. Hyperplasia, keratosis, dysplasia, and carcinoma in situ of the vocal cords -- a follow-up study. Clin Otolaryngol 1982; 7:11-27.
Hintz BL, Kagon AR, Nussbaum H, et al. A "watchful waiting" policy for in situ carcinoma of the vocal cords. Arch Otolaryngol 1981; 107; 746-751.
Hojslet PE, Nielsen VM, Palvio D. Premalignant lesions of the larynx. A follow-up study. Acta Otolaryngol (Stockh) 1989; 107:150-155.
Horiuchi K, Mishima K, Ichijima K, et al. Epstein-Barr virus in the proliferative diseases of squamous epithelium in the oral cavity. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995; 79:57-63.
Huang WY, Coltrera M, Schubert M, et al. Histopathologic evaluation of proliferating cell nuclear antigen (PC10) in oral epithelial hyperplasias and premalignant lesions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1994; 78:748-754.
Ikeda N, Downer MC, Ishii T, et al. Annual screening for oral cancer and precancer by invitation to 60-year-old residents of a city in Japan. Community Dent Heath 1995; 12:133-137.
Kambic V. Difficulties in management of vocal cord precancerous lesions. J Laryngol Otol 1978; 92:305-315.
Kinoshita Y, Inoue S, Honma Y, et al. Diagnostic significance of nuclear DNA content and nuclear area in oral hyperplasia, dysplasia, and carcinoma. J Oral Maxillofac Surg 1992; 50:728-733.
Kleinsasser O. Uber den Krankheitsverlauf bei Epithelhyperplasien der Kehlkopfscheimaut und die Entstehung von Karzinomen: IV. Mitteilung. Z Laryngol Rhinol Otol 1963; 42:541-558.
Kramer IRH. Carcinoma in situ of the oral mucosa. Int Dent J 1973; 23:94-99.
Kratochvil FJ, Cioffi GA, Auclair PL, et al. Virus-associated dysplasia (bowenoid papulosis?) of the oral cavity. Oral Surg Oral Med Oral Pathol 1989; 68:312-316.Lay KM, Sein K, Myint A, et al. Epidemiologic study of 600 villagers of oral precancerous lesion in Bilugyan: preliminary report. Community Dent Oral Epidemiol 1982; 10:152-155.
Liu SC, Sauter ER, Clapper ML, et al. Markers of cell proliferation in normal epithelia and dysplastic leukoplakias of the oral cavity. Cancer Epidemiol Biomarkers Prev 1998; 7:597-603.
Lumerman H, Freedman P, Kerpel S. Oral epithelial dysplasia and the development of invasive squamous cell carcinoma. Oral Surg Oral Med Oral Pathol 1995; 79:321-329.
Manzone H, Billings PC, Cummings WN, et al. Levels of proteolytic activities as intermediate marker endpoints in oral carcinogenesis. Cancer Epidemiol Biomarkers Prev 1995; 4:521-527.
Martin IC, Kerawala CJ, Reed M. The application of toluidine blue as a diagnostic adjunct in the detection of epithelial dysplasia. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 85:444-446.
Mashberg A. Erythroplasia vs leukoplakia in the diagnosis of early asymptomatic oral squamous cell carcinoma. N Engl J Med 1977; 297:109-110.
Mashberg A, Morrissey JB, Garfinkel L. A study of the appearance of early asymptomatic oral squamous cell carcinoma. Cancer 1973; 32:1436-1445.
Mashberg A, Samit AM. Early detection, diagnosis, and management of oral and oropharyngeal cancer. CA 1989; 39:67-88.
Matzow T, Boysen M, Kalantari M, et al. Low detection rate of HPV in oral and laryngeal carcinomas. Acta Oncol 1998; 37:73-76.
McCance DJ, Kopan R, Fuchs E, et al. Human papillomavirus type 16 alters human epithelial cell differentiation in vitro. Proc Natl Acad Sci USA 1988; 85:7169-7173.
McGavran MH, Bauer WC, Ogura JH. Isolated laryngeal keratosis--its relation to carcinoma of the larynx based on a clinicopathologic study of 87 consecutive cases. Laryngoscope 1960; 70:932-951.
Merrell P, Carpenter W, Silverman S Jr, et al. Reactivity of monoclonal antibodies 17.13 and 63.12 with oral epithelial dysplasia and hyperkeratosis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997; 83:367-372.
Miller AH. Carcinoma-in-situ of the larynx: a 20 year study of the results of management. Am J Surg 1970; 120:492-494.
Miller AH, Fisher HR. Clues to the life history of carcinoma in situ of the larynx. Laryngoscope 1971; 81:1475-1480.
Miller CS, White DK. Human papillomavirus expression in oral mucosa, premalignant conditions, and squamous cell carcinoma: a retrospective review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997; 82:57-68.
Mincer HH, Coleman SA, Hopkins KP. Observations on the clinical characteristics of oral lesions showing histologic epithelial dysplasia. Oral Surg Oral Med Oral Pathol 1972; 33:389-399.
Myerson MC. Smoker's larynx: clinical pathologic entity. Ann Otol Rhin Laryngol 1950; 59:541-546.
Neville B, Damm D, Allen C, et al. Oral and maxillofacial pathology. Philadelphia: W. B. Saunders, 1995: 284-288.
Nielsen H, Norrild B, Vedtofte P, et al. Human papillomavirus in oral premalignant lesions. Eur J Cancer B Oral Oncol 1996; 32B:264-270.57.
Norris CM, Peale AR. Keratosis of the larynx. J Laryngol Otol 1973; 77:635-647.
Partridge M, Emilion G, Pateromichelakis S, et al. Allelic imbalance at chromosomal loci implicated in the pathogenesis of oral precancer, cumulative loss and its relationship with progression to cancer. Oral Oncol 1998; 34:77-83.
Penned F, Fletcher GH. Results in irradiation of the in situ carcinoma of the vocal cords. Cancer 1976; 37:2586-2590.
Pindborg JJ, Daftary DK, Mehta FS. A follow-study of 61 oral dysplastic precancerous lesions in Indian villagers. Oral Surg Oral Med Oral Pathol 1977; 43:383-390.
Prout MN, Sidari JN, Witzburg RA, et al. Head and neck cancer screening among 4611 tobacco users older than 40 years. Otolaryngol Head Neck Surg 1997; 116:201-208.
Putney FJ, O'Keefe JJ. Clinical significance of keratosis of the larynx as a premalignant lesion. Ann Otol Rhin Laryngol 1953; 62:348-357.
Queyrat L. Erythroplasie du gland. Bull Soc Fr Dermatol Syphiligr 1911; 22:378-382.
Reibel J, Clausen H, Dabelsteen E. Staining patterns of human pre-malignant oral epithelium and squamous cell carcinomas by monoclonal anti-keratin antibodies. Acta Pathol Microbiol Immunol Scand (A) 1985; 93:323-330.
Rubin IC. The pathologic diagnosis of incipient carcinoma of the uterus. Am J Obstet 1910; 62:668-676.
Santa Cruz DJ. Skin. In: Henson DE, Albores-Saavedra J. The pathology of incipient neoplasia. Philadelphia: W. B. Saunders, 1986:5-16.
Schiller W. Untersuchungen zur Entstehung der Geschwulste; Collumcarcinom des Uterus. Virchows Arch f path Anat 1927; 263:279-367 .
Scully C, Beyli M, Feirrero M, Ficarra G, et al. Update on oral lichen planus: aetiopathogenesis and management. Crit Rev Oral Bio Med 1998; 9:86-122.
Shafer WG. Oral carcinoma in situ. Oral Surg Oral Med Oral Pathol 1975; 39:227-238.
Shafer WG, Hine MK, Levy BM. A textbook of oral pathology, 3rd edition. Philadelphia: W. B. Saunders, 1972:106-108.
Shafer WG, Waldron CA. Erythroplakia of the oral cavity. Cancer 1975; 36:1021-1028.
Shedd DP, Hukill PB, Kigerman MM, et al. A clinicopathologic study of oral carcinoma in situ. Am J Surg 1963; 106:791-796.
Silverman S Jr. Oral Cancer, 4th ed. Atlanta, Georgia: American Cancer Society, 1998.
Silverman S Jr, Bahl S. oral lichen planus update: clinical characteristics, treatment responses, and malignant transformation. Am J Dent 1997; 10:259-263.
Silverman S Jr, Gorsky M, Lozada F. Oral leukoplakia and malignant transformation: a follow-up study of 257 patients. Cancer 1984; 53:563-568.
Slaughter DP, Southwick HW, Smejkal W. "Field cancerization" in oral stratified squamous epithelium: clinical implications of multicentric origin. Cancer 1953; 6:963-968.
Speight PM, Farthing PM, Bouquot JE. The pathology of oral cancer and precancer. Curr Diag Pathol 1997; 3:165-167.
Tiitta O, Happonen RP, Virtanen I, et al. Distribution of tenascin in oral premalignant lesions and squamous cell carcinoma. J Oral Pathol Med 1994; 23:446-450.
Waldron CA. Oral epithelial tumors. In: Gorlin RJ, Goldman HM (eds). Thoma's Oral pathology, 6th ed. St. Louis: C. V. Mosby; 1970:824-826.
Vedtofte P, Holmstrup P, Hjorting-Hansen E, et al. Surgical treatment of premalignant lesions of the oral mucosa. Int J Oral Maxillofac Surg 1987; 16:656-664.
Waldron CA, Shafer WG. Leukoplakia revisited: a clinicopathologic study of 3256 oral leukoplakias. Cancer 1975; 36:1386-1392.
WHO Collaborating Centre for Oral Precancerous Lesions. Definitions of leukoplakia and related lesions: an aid to studies on oral precancers. Oral Surg Oral Med Oral Pathol 1978; 46:518-539.
Wright A, Shear M. Epithelial dysplasia immediately adjacent to oral squamous cell carcinoma. J Oral Pathol 1985; 14:559-564.
Zain RB, Ideda N, Razak IA, et al. A national epidemiological survey of oral mucosal lesions in Malaysia. Community Dent Oral Epidemiol 1997; 25:377-383.
Zhang L, Xiao Y, Priddy R. Increase in placental glutathione S-transferase in human oral epithelial dysplastic lesions and squamous cell carcinomas. J Oral Pathol Med 1994; 23:75-79.
Table 1: Estimated relative malignant transformation potential of precancerous lesions (clinical terms only) of the oral, pharyngeal and laryngeal mucosa. Modified from Speight et al, Curr Diag Pathol 1997; 3:165-167.3
# reverse smoking: smoking with the
lit end of the cigarette in one's mouth
Table 2: Malignant transformation rates (%) for oral CIS and/or severe epithelial dysplasia, listed by year of publication. Lesions appeared clinically as red, white or combined red and white macules, i.e. not all were pure erythroplakias. Cancers not arising from the site of the precancer are excluded.
* sum total of time which
lesions were collectively followed (mean follow-up times are in parentheses)
Table 3: Malignant transformation rates (%) for laryngeal CIS and/or severe dysplasia, as determined by all English language follow-up studies published to date, listed by year of publication.
* includes only cancers
developing in the anatomic area of the previously recorded dysplastic lesions
Table 4: Molecular and other markers of potential significance to oral epithelial dysplasia & CIS.
* identified as "alcoholic,"
"heavy drinker," or with a "drinking problem" in medical