Little is known about the genetic regulation of medulloblastoma dissemination, but metastatic medulloblastoma is highly associated with poor outcome. We obtained expression profiles of 23 primary medulloblastomas clinically designated as either metastatic (M+) or non-metastatic (M0) and identified 85 genes whose expression differed significantly between classes. Using a class prediction algorithm based on these genes and a leave-one-out approach, we assigned sample class to these tumors (M+ or M0) with 72% accuracy and to four additional independent tumors with 100% accuracy. We also assigned the metastatic medulloblastoma cell line Daoy to the metastatic class. Notably, platelet-derived growth factor receptor alpha (PDGFRA) and members of the downstream RAS/mitogen-activated protein kinase (MAPK) signal transduction pathway are upregulated in M+ tumors. Immunohistochemical validation on an independent set of tumors shows significant overexpression of PDGFRA in M+ tumors compared to M0 tumors. Using in vitro assays, we show that platelet-derived growth factor alpha (PDGFA) enhances medulloblastoma migration and increases downstream MAP2K1 (MEK1), MAP2K2 (MEK2), MAPK1 (p42 MAPK) and MAPK3 (p44 MAPK) phosphorylation in a dose-dependent manner. Neutralizing antibodies to PDGFRA blocks MAP2K1, MAP2K2 and MAPK1/3 phosphorylation, whereas U0126, a highly specific inhibitor of MAP2K1 and MAP2K2, also blocks MAPK1/3. Both inhibit migration and prevent PDGFA-stimulated migration. These results provide the first insight into the genetic regulation of medulloblastoma metastasis and are the first to suggest a role for PDGFRA and the RAS/MAPK signaling pathway in medulloblastoma metastasis. Inhibitors of PDGFRA and RAS proteins should therefore be considered for investigation as possible novel therapeutic strategies against medulloblastoma.

The best operative intervention for children with arachnoid cysts remains the subject of controversy. Recent reports stress that craniotomy for cyst fenestration is associated with a low incidence of morbidity and mortality and may leave the child shunt-independent. The cases of 40 pediatric patients with arachnoid cysts treated between 1978 and 1989 are reported. Five children with mild symptoms and small cysts that remained stable on follow-up studies have not required surgical intervention. Of 15 patients with cysts initially treated by fenestration, 10 (67%) showed no clinical or radiographic improvement postoperatively and have undergone cyst-peritoneal (eight patients) or ventriculoperitoneal (VP) shunting (one patient), or revision of a VP shunt placed for hydrocephalus before cyst fenestration (one patient). Two other patients with existing VP shunts required no further procedures. Thus, only three (20%) of 15 patients initially treated by fenestration remain shunt-independent after a median follow-up period of 8 years. The 20 other patients were initially treated by cysts shunting and all improved postoperatively; shunt revision has been necessary in six (30%) of these 20 patients because of cysts recurrence. Cyst location influenced the success of shunt treatment; none of the seven middle cranial fossa cysts treated by shunting have required revision, but results with cysts in other locations were less favorable. In all locations, though, shunting was more successful than fenestration. It is concluded that cyst-peritoneal or cyst-VP shunting is the procedure of choice for arachnoid cysts in most locations, including those in the middle cranial fossa.

Little is known about the risk of developing multicentric disease in patients with juvenile pilocytic astrocytoma (JPA), and even less about its prognosis. Only five cases have been reported. Between 1986 and 1992, the authors treated 90 patients with either primary or recurrent JPA, 11 of whom developed multicentric spread. Ten patients had primary tumors in the hypothalamic region, eight were under 4 years of age at initial diagnosis, all had initially undergone a subtotal resection or biopsy, and 10 received postoperative multiagent chemotherapy or irradiation for residual disease. Multicentric spread was discovered immediately to 108 months after initial diagnosis; nine patients were asymptomatic at the time. Most patients received chemotherapy for the multicentric disease, which was found throughout the craniospinal axis. During 21 to 148 months of follow-up monitoring, seven patients had stabilization or regression of multicentric disease and four died. Patients with hypothalamic region tumors were 23 times more likely to develop multicentric spread than were those with primary tumors located elsewhere (p < 0.001). Based on this review, it is concluded that multicentric spread of JPA occurs more frequently than was previously recognized. In patients with subtotally resected JPA and several years of follow-up review via magnetic resonance imaging, the incidence of recurrence in a site different from the original was 12%. Patients with subtotally resected JPA in the hypothalamic region should be considered to be at high risk for developing multicentric spread. Chemotherapy appears useful in stabilizing multicentric disease. Earlier detection and intervention may result in longer disease-free survival in patients with multicentric spread of JPA.

Inactivation of the PTCH tumor suppressor gene occurs in a subset of sporadic medulloblastomas, suggesting that alterations in the PTCH pathway may be important in the development of this tumor. In order to address the frequency of genetic alterations affecting genes in this pathway, we used a combination of loss of heterozygosity (LOH) analysis, single‐stranded conformational polymorphism (SSCP) analysis, and direct sequencing of DNA samples from sporadic primitive neuroectodermal tumors (PNETs). To identify alterations in the PTCH gene, we performed LOH analysis on 37 tumor DNA samples. Of those with matched constitutional DNA samples, one demonstrated LOH. Of those without matched constitutional DNA, six were homozygous with all markers. All exons of the PTCH gene were sequenced in these seven tumors, and three mutations were found. To identify alterations in the SHH and SMO genes, we analyzed all exons of both genes in 24 tumors with SSCP and sequenced any exons that showed aberrant band patterns. No mutations were found in either SHH or SMO in any tumor. We also identified the following genes as candidate tumor suppressors based on their roles in controlling hh/ptc signaling in Drosophila: EN‐1 and EN‐2, deletion of which results in a lack of cerebellar development in mice; SMAD family members 1–7, and protein kinase A subunits RIα, RIβ, RIIβ, Cα, and Cβ. Each of these genes was investigated in a panel of 24 matched constitutional and tumor DNA samples. Our search revealed no mutations in any of these genes. Thus, PTCH is the only gene in this complex pathway that is mutated with notable frequency in PNET. Genes Chromosomes Cancer 27:44–51, 2000. © 2000 Wiley‐Liss, Inc.

Neurofibromatosis 1 (NF-1) (von Rekhlinghausen disease) and neurofibromatosis 2 (NF-2) (bilateral acoustic neurofibromatosis) have been recently recognized to be distinct disorders through genetic linkages. The authors compared the cranial magnetic resonance (MR) images of 53 patients with NF-1 and 11 with NF-2. In the NF-1 group, MR imaging revealed 19 patients with optic gliomas and eight with parenchymal gliomas. In 32 patients, foci of prolonged T2, similar to those reported previously as hamartomas, were identified in the cerebellar peduncles, globus pallidus, midbrain, and other locations. The frequency of these foci was related to both age and the presence of optic gliomas. In the NF-2 group, MR imaging revealed eight patients with cranial nerve schwannomas and six with meningiomas (in addition to acoustic schwannomas in all 11). These findings demonstrate that NF-1 and NF-2 are different diseases requiring different imaging protocols. NF-1 seems to be associated with tumors of astrocytes and neurons and NF-2 with tumors of meninges and Schwann cells.