Page 192 - Strategies for non-invasive managementof high-grade cervical intraepithelial neoplasia - prognostic biomarkers and immunotherapy Margot Maria Koeneman

Chapter 10
based on conventional pathological criteria, show spontaneous regression rates for CIN2 of 40- 74%, compared to 22-33% for CIN3 lesions.[1, 20] Other studies confirm the high regression rate of CIN2 lesions.[31] Based on these results, guidelines now advice upon observational management of CIN2 in younger women, to reduce overtreatment and associated long-term obstetric complications.[32] In clinical histopathological assessment, however, the distinction between CIN2 and CIN3 lesions is challenging. Studies show that CIN2 is a much less reproducible diagnosis than CIN3 and distinction between CIN2 and CIN3 is not always clear.[33-36] Concordance in CIN2 diagnosis between different pathologists was found in only 19-53% of cases, compared to 48-62% in CIN3 lesions. The inconsistency in CIN2 diagnosis raises discussion as to whether CIN 2 actually corresponds to a well-defined phase of the pathogenetic pathway of infection and transformation in CIN. If CIN2 is considered not to be individual entity, the histological image of CIN2 may represent the late stage of a productive infection or the early stage of a transforming infection, which are considered two biologically different phases in the pathogenetic pathway of infection and transformation.[37] Although the histological image of these two forms of CIN2 may look similar, their biological behavior is markedly different. The high regression rate of CIN2 may then be explained by the fact that a substantial subset of CIN2 lesions represents a productive infection rather than a transforming infection. A solution to this discussion would be a strategy in which the lesion is not classified according to CIN grade, but according to its potential for spontaneous regression. After all, that is the actual clinical relevance of the diagnosis. This brings us back to the importance of prognostic biomarkers in high-grade CIN. While prognostic biomarkers are lacking, other methods should be applied to distinguish between CIN2 and CIN3. Uleberg et al. studied 114 water-soluble proteins in supernatants from 20 fresh cervical biopsies (10 CIN2 and 10 CIN3 lesions).[38] The histological diagnosis was based on p16 and Ki67 staining, but no information was provided on HPV status. Cytokeratin (CK)2 was found to be a strong discriminator between CIN2 and CIN3, with 90% overall correct classifications. CK2 expression was expressed in the low and middle part of the epithelium in CIN2 lesions and was not expressed in CIN3 lesions. A difference in CK2 expression seems biologically plausible. This cytokeratin is normally expressed in late differentiation of the epidermal layers of normal skin. Its varied expression in CIN indicates that epithelial cells in CIN2 have a greater tendency towards differentiation.[38] These diagnostic properties of CK2 have not been validated in larger studies and have therefore not been implemented in clinical practice. Until the development of better methods, we advocate the distinction of CIN2 and CIN3 lesions based on conventional histopathological criteria, including p16 and Ki67 staining, with the aim to provide young women an opportunity for conservative management in case of CIN2.
In the era of personalized medicine and shared decision-making, individual prediction of disease outcome in CIN2 would enhance patient counseling with regard to management options. We therefore developed a prediction model for spontaneous regression of CIN2. Since it consists of simple and inexpensive clinical and pathological parameters, it is widely applicable. The model, based on smoking status, PAP smear outcome, concomitant CIN1 and number of biopsies containing CIN2, has a reasonable discriminative capacity and is accurate, especially in the prediction of disease regression. Upon external validation, the model could function as a tool for a more individualized approach to CIN2 management, in a population with mixed hrHPV positive
190