Oligoflendrogliomas: Clinicopathological correlations

Since the original description of glial neoplasms consisting of cells with oligodendroglial morphology [1] many papers have addressed the clinical behavior of this tumor. While their name was derived from a characteristic appearance in silver stains as tumors composed of glial cells with few and short processes, the hallmark of oligodendrogliomas as shown in hematoxylin and eosin stains became the typical honeycomb structure. Soon after its recognition it became clear that in a considerable number of oligodendrogliomas astrocytic cells were encountered , and after the introduction of immuno-histochemical stains for glial fibrillary acidic protein , a marker for the astrocytic cell lineage, even more oligodendroglial neoplasms appeared to harbor astrocytic cells [2, 3]. Only a limited number of recent studies specifically addresses the problem of how to distinguish oligodendrogliomas from the mixed oligo-astrocytomas by mentioning maximum allowable percentages of astrocytic cells [6-9]. In recent years immunohistochemical markers specifically reacting with the cells of oligodendrogliomas have been sought but not yet found [4, 5]. Besides the well known difficulties in predicting clinical outcome of neoplasms based on histopath-ology, as interference of other prognostically relevant parameters, and besides the limited representation of a small biopsy sample from tumors as het-erogenous as gliomas [10], clinicopathological investigations were hampered by the relatively rareness of this member of the glioma family. Furthermore , there is no consensus regarding the legiti-mation of individual histopathological features used in grading schemes for oligodendrogliomas. In several recent studies grading schemes consisting of various features have been tested with regard to their predictive power [6, 7, 9, 11, 12]. No major differences were apparent between the survival curves resulting from application of the original, rather ill-defined scheme of Kernohan, and those obtained by using the schemes later developed by the AFIR or that used by a french-american group (St. Anne-Mayo) [9, 12]. Some studies were aimed at identifying individual histopathological features with significance for prognosis. The results of these studies do not provide a consensus. In a large Norwegian study the features cell density, necrosis and microcysts were found to have prognostic significance [7]. In contrast , in the largest study on oligodendrogliomas to date so far, viz. the AFIP series, only nuclear pleo-morphism was identified as individual significant factor [6]. Stepwise regression analysis used in another American study revealed that necrosis and the number of mitoses contained all prognostically usefull information. The St. Anne-Mayo scheme, not exclusively used for …