gms | German Medical Science

Objective: Human meningiomas are mostly benign, but they may relapse or are not fully resectable because of anatomical restraints. So far, alternative systemic chemotherapeutical treatment has been largely disappointing. In experimental research, mostly aggressive types of meningiomas or genetically manipulated celllines are used. Our goal was to test whether primary cell cultures of benign human meningiomas would show histomorphological and immunohistochemical characteristics of the original human meningioma after intracranial implantation into nude mice.

Methods: We used primary cell cultures of 11 benign human meningiomas, which were passaged 6 times. A cell suspension of each human meningioma was then stereotactically implanted into the prefrontal cortex of 2-5 NMRI nu/nu mice. After 3 months, the mice were sacrificed and their brains paraffin-embedded. Both mouse brains and human meningiomas were cut into 5 µm slices. One series was H&E stained for morphological characterization and measurement of the tumor volume. The other series were immunohistochemically processed for von Willebrand Factor to assess the intratumoral microvessel density (iMD), and for Ki-67 to analyze the proliferation index. In addition, the extent of expression and intensity of epithelial membrane antigen (EMA) was evaluated by using an immunhistochemical score (IHS). Since cyclooxygenase-2 (COX-2) is a possible target for pharmaceutical interventions, we also evaluated COX-2 expression.

Results: In 93% of the patients, tumor formation was observed after implantation of primary human meningioma cells. These tumors were generally small, but with meningothelial morphology comparable to the original human tumors. The proliferation index (i.e., the percentage of positive Ki-67 stained tumor cells) and the COX-2 IHS did not differ between mouse and original human tumors. Instead, the EMA IHS and the iMD was reduced in murine tumors (both p<0.009). Besides, in mice, the tumor volume correlated positively with the iMD (p=0.013). Nevertheless, antigen-expression of meningioma from individual human specimen was variable in different mice, which seems to reflect a tumor-host-interaction.

Conclusions: We have shown here that implantation of benign human meningioma primary cell cultures into mice results in tumor formation with morphological and immunohistological features comparable to the original human tumor. This rodent meningioma model may therefore be suitable to test novel therapeutic agents.