How far can we trust cell cycle biomarkers for the identification of high‐risk melanoma patients?
T. Simonart,M. Heenen
DOI: https://doi.org/10.1111/j.0303-6987.2005.00372.x
2005-05-01
Journal of Cutaneous Pathology
Abstract:To the Editor, Li et al. recently reported that loss of p27 protein, a cell cycle regulator that acts as a cyclindependent kinase inhibitor, is implicated in melanoma progression. Low expression of this protein significantly correlated with American Joint Committee on Cancer (AJCC) stage but, however, it did not impact significantly on prognosis. Although we appreciate Dr Li and colleagues’ efforts to provide additional biomarkers for melanoma, we remain sceptical of the reported use of p27 to increase accurate identification of high-risk patients. Cell cycle biomarkers have been found to have prognostic significance in a variety of malignant neoplasms and are used routinely in order to make treatment decisions. However, despite intensive research and numerous studies over the past 15 years, we are still in the uncertainty regarding the application of measures of cell cycle markers as prognostic factors for cutaneous melanoma. By definition, a useful prognostic factor needs to detect, at an early stage, patients who are at risk to develop a metastatic disease. In melanoma, tumour thickness remains the strongest and more used indicator of the risk of dissemination of the primitive tumour. However, a small but significant proportion of patients with thin melanoma may develop a metastatic disease with a fatal outcome. Conversely, some patients with a thick tumour sometimes may exhibit a long disease-free survival, highlighting the need of prognostic indicators more powerful than tumour thickness. Moreover, while several studies found that the expression of some cell cycle markers, such as Ki-67, correlates with the development of metastases in thin melanomas, other studies fail to demonstrate such a link. Analogously, there are, in addition, contradictory results about p27 whose expression has been showed to be upregulated in some studies on invasive melanoma and melanoma cell lines, which renders the clinical dermatopathologist completely blind to the problem. Therefore, although Li et al. reported a correlation between loss of p27 and melanoma progression, a significant proportion of metastatic melanomas showed an increase in p27 expression. As outlined by the authors, these discrepancies could be related to tumour microenvironment and, more particularly, to post-transcriptional/post-translational p27 upregulation caused by tumour hypoxia. Because hypoxia is a spatially and temporally very heterogeneous phenomenon resulting from stochastic tumour and host tissue-specific processes, we wonder whether studies involving cell cycle control regulators should not be limited only to the invasive fronts of tumours, as previously suggested for other tumours.