CIN III lesions and regression: retrospective analysis of 635 cases

BACKGROUND The rate of spontaneous regression in CIN III lesions is controversial. Whereas some studies have reported high regression rates of up to 38% after prolonged biopsy-conus intervals, others have shown rates between 0 and 4% without considering time intervals. Identification of young patients with potentially regressing CIN III could offer the chance to avoid conisation, thus lowering the risk of preterm labour. METHODS To further clarify the facts, we retrospectively compared 635 biopsies showing CIN III with the diagnosis of the conisation. Either regression (CIN I or less) or non-regression (CIN II and higher) was recorded. Diagnoses were made by light microscopy and p16 immunostaining. RESULTS Conisation was performed between 2 and 463 days after biopsy (median 8.9 weeks). Six hundred twenty one (98%) were HPV-HR positive. In 345 cases, HPV subtyping was available, showing HPV16 infection in 57%. Routine processing of the conisation tissue showed no corresponding CIN lesion (< CIN II) in 40 cases (6.3%). Additional step sectioning of the tissue revealed small CIN II+ lesions in 80%. Finally, eight cases (1.3%) fulfilled the criteria of regression. No regression was seen in HPV16 positive cases. Twelve invasive carcinomas were detected by routine processing of the conisation tissue. CONCLUSION These results are in contrast with some prior reports that might have overestimated spontaneous regression of CIN III. Study size and an accurate discrimination between CIN II and CIN III lesions by histopathology seem to be the most likely factors to explain the diverging results published. Complete step sectioning of the whole tissue is also mandatory in questionable cases. Although theories exist that the initial biopsy might stimulate the immune system, thus triggering regression within weeks, our data do not substantially support such a mechanism. Overall, the chance of a CIN III lesion to regress rapidly within weeks or months after diagnosis seems to be small. We found more previously undetected invasive cancer than we observed regression. Therefore, a change in the current policy to treat CIN III lesions is unwarranted.