Most anatomic pathologists are familiar with a multihit model for the development of epithelial malignancies. As such, they are accustomed to diagnosing noninvasive epithelial precursors at a number of anatomic sites. The best studied model for the neogenesis of squamous cell carcinoma is the uterine cervix, and most anatomic pathologists are well versed in diagnosing squamous intraepithelial lesions in this location.
There is also abundant evidence that squamous cell carcinoma of the upper aerodigestive tract develops through worsening intraepithelial neoplasia. Clinically, precursor lesions may appear white or red (or speckled) and are termed, respectively, leukoplakia or erythroplakia. These clinical designations do not directly correspond to definitive histologic diagnoses, although speckled and red lesions (erythroplakia) are more likely to show significant squamous dysplasia and be associated with concurrent or subsequent squamous cell carcinoma.
1,2 and
3 The data regarding the risk of these lesions progressing to invasive malignancy are varied (
Table 3.1), although worsening degrees of dysplasia are associated with greater risks in most studies.
3,4,5,6,7,8,9,10,11,12,13,14,15,16,17 and
18
Unlike in the cervix, where high-risk human papillomavirus (HPV) infection causes the vast majority of intraepithelial lesions, there are a myriad of causes of upper aerodigestive tract squamous intraepithelial lesions, and HPV is involved in the development of only a minority of cases.
19 In addition, cervical dysplasias arise from a histologically distinct transformation zone that does not have an equivalent in the aerodigestive tract. Thus, the histologic features of most squamous dysplasias of the upper aerodigestive tract are not identical to those of the cervix, although obvious similarities exist. Undoubtedly, there is a complete spectrum of histologic changes that can be seen between the normal squamous epithelium of the upper aerodigestive tract and squamous cell carcinoma. How many lines one wishes to draw through this continuum is subjective. For the purpose of this chapter, we shall use the 2005 WHO classification system (
Table 3.2), although we believe a five-tiered system may, among other things, lead to some considerable difficulties with diagnostic reproducibility.
20