Combination chemotherapy with or without radiotherapy has had only modest efficacy in the treatment of primary CNS lymphoma. Median survival of these patients, treated primarily with radiotherapy, is 13 months; 5-year survival is less than 5%. Thirty consecutive non-acquired immune deficiency syndrome patients with primary CNS lymphoma were treated with barrier-dependent chemotherapy using intraarterial mannitol to open the blood-brain barrier (BBB). Follow-up included extensive neuropsychologic testing of all patients. Thirteen patients received cranial radiation 1 to 9 months before referral (group 1). Seventeen patients received initial BBB disruption chemotherapy with subsequent radiation only for tumor progression or recurrence (group 2). The difference in median survivals from diagnosis--17.8 months for group 1 and 44.5 months for group 2--was statistically significant (P = .039). Group 1 survival is comparable with the 20-month median survival of a historical series of patients (n = 208) treated with radiotherapy with or without chemotherapy. Group 2 patient survival represents an advance in the survival of CNS lymphoma and was associated with preservation of cognitive function in six of seven nonirradiated complete responders observed for 1 to 7 years. Patient toxicity was manageable in this intensive therapeutic regimen. In this series, a plateau in survival curves suggests that a major portion of these patients may be cured without the neuropsychologic sequelae associated with cranial radiation.

No abstract

CARBOPLATIN AND ETOPOSIDE have been investigated in preclinical studies and a limited toxicity study in 13 patients; these studies have established carboplatin and etoposide as a tolerable combination when administered with blood-brain barrier disruption. The studies also found a predictable dose-limiting toxicity of myelosuppression. Subsequently, a broad efficacy trial of this regimen was carried out. A total of 34 patients, ranging in age from 7 to 72 years, underwent a combination chemotherapy regimen of carboplatin (200 mg/m2 administered intra-arterially) and etoposide (200 mg/m2 administered intravenously) administered with blood-brain barrier disruption on each of 2 consecutive days every 28 days. The diagnoses included glioblastoma multiforme (n = 3), malignant astrocytoma (n = 8), malignant astrocytoma-oligodendroglioma (n = 1), primitive neuroectodermal tumor (n = 4), disseminated germ cell tumor of the central nervous system (CNS) (n = 6), CNS lymphoma (n = 7), and metastatic carcinoma (n = 5). The major toxicity observed in patients treated with multiple courses of this regimen was the expected reversible myelosuppression and an unexpected, irreversible high-frequency hearing loss. Of these 34 patients, 22 had measurable disease, and 9 radiographic responses (50% or more decrease in enhancing tumors) were observed in these patients. Carboplatin and etoposide with blood-brain barrier disruption is an active regimen in the treatment of malignant astrocytomas and has shown dramatic responses in primitive neuroectodermal tumors and CNS lymphoma. Additionally, the durability of responses in patients with disseminated CNS germ cell tumors is encouraging. However, such therapy is associated with unexpected high-frequency hearing loss; even so, on the basis of the favorable responses in patients with primitive neuroectodermal tumors, germ cell tumors, and lymphomas, the study of this regimen for those tumors is being extended in a multiinstitutional trial that also includes cytoxan to further evaluate the potential enhanced drug delivery.