Background: Lesions that are histologically classified as precursors of cervical cancer, which are often referred to as squamous intraepithelial lesions (SILs), represent a heterogeneous clinical entity that can be associated with many different types of human papillomaviruses (HPVs) and have a variable biologic behavior. Approximately one half of low-grade SILs behave as non-neoplastic, productive viral lesions that frequently regress spontaneously, whereas the other half behave as neoplasms and either persist or progress to a histologically higher grade lesion. Identification of biomarkers that reliably differentiate those low-grade SILs with the properties of a non-neoplastic viral infection from those with the properties of neoplasia would provide a more rational basis for decisions about disease management. Since monoclonality is a hallmark of neoplasia irrespective or organ site, clonal status might represent one such biomarker.
Purpose: To better understand the pathobiology of SILs, we analyzed the clonality of low-grade and high-grade SILs and compared their clonal status with their associated HPV types.
Methods: One hundred forty formalin-fixed, paraffin-embedded cervical biospy and loop electrosurgical specimens, originally diagnosed as SILs, were obtained from the pathology archives of both the Columbia-Presbyterian Medical Center and Kyto Diagnostics in New York. Clonality was determined with the use of a polymerase chain reaction (PCR) based method that detects nonrandom X-chromosome inactivation. This PCR-method amplifies a polymorphic region of the androgen receptor gene that is flanked by several differentially methylated enzyme sites. The same tissue was also analyzed for HPV DNA with the use of PCR and both L1 and E6 "consensus" primers.
Results: All 25 evaluable cases of high-grade SILs were determined to be monoclonal. Although 54 (68%) of 79 evaluable low-grade SILs were monoclonal, 25 (32%) of 79 low-grade SILs were polyclonal. A strong association was observed between HPV type and clonal status, with a total of 71 (47 low-grade and 24 high-grade) SILs determined to be monoclonal and containing HPV types 16, 18, 31, 33, 35, 39, 45, 56, 58, or 65. In contrast, 22 (92%) of the 24 low-grade SILs that contained another type of HPV were polyclonal (Fisher's exact test, two-sided, P</-.001).
Conclusions: Our findings suggest that the histopathologic entity termed low-grade SIL consists of two different types of lesions that are biologically distinct. One lesion is monoclonal and is associated with HPV types 16, 18, 31, 33, 35, 39, 45, 56, 58, or 65. The second types of low-grade SIL is polyclonal and is associated with other types of HPV.