Cellular schwannoma. A clinicopathologic, DNA flow cytometric, and proliferation marker study of 70 patients.

Abstract

A clinicopathologic study of 70 cases of cellular schwannoma was performed to assess their distribution, response to therapy, and rate of recurrence relative to modern prognostic indicators. Seventy-one cellular schwannomas from 70 patients were retrieved from the files of the Mayo Clinic Tissue Registry. The significance of mitotic index, proliferative marker staining (proliferating cell nuclear antigen and MIB1), immunochemical p53 expression, and DNA ploidy were assessed relative to tumor behavior, particularly recurrence. All parameters were subject to statistical analysis (Student's t test). Cellular schwannomas represented 4.6% of benign peripheral nerve tumors operated on at the Mayo Clinic. In 15% of the cases, an initial diagnosis of malignancy had been made. The median patient age was 47.7 years (range, 15-80 years) and the female-to-male ratio was 1.6:1. The principle tumor locations were the para- and intraspinal regions, including the sacrum (64%), extremities (25%), and intracranial space (8%). All tumors consisted primarily of hypercellular, compact, Antoni A tissue. Mitoses (< or = 4/10 hPF]) were observed in 71% of the cases. Foci of necrosis were noted in 11% of cases. Ultrastructural studies and immunohistochemistry clearly demonstrated features of schwannian differentiation. Surgery was the treatment in all cases. Excision was intralesional to gross total in the majority; total resection with wide margins was undertaken in three tumors, each of which had initially been considered malignant. Follow-up in 47 patients (67%) ranged from 1 to 29 years (mean, 7.7 years) and revealed recurrences in 11 patients (23.4%): no patient experienced metastasis or died of tumor. Although no correlation existed between recurrence and DNA ploidy, percent S-phase determinations, proliferation marker (PCNA, MIB1) staining, or the frequency of p53 immunoreactivity, a statistically significant correlation (P < 0.001) was observed, however, between recurrence and mitotic indices. Proliferation indices, as defined by immunochemical analysis, are not useful predictors of recurrence in cellular schwannoma. In lesions not completely resected, tumor recurrence is significantly correlated with mitotic count. The significant overall frequency of recurrence in this series is attributable to a high proportion of intraspinal and intracranial tumors. Our study confirms the benign nature of cellular schwannoma and underscores the necessity of distinguishing them from malignant peripheral nerve sheath tumors, lesions that often require adjuvant therapy.