INTRODUCTION — Medulloblastoma is the most common malignant brain tumor of childhood. It accounts for approximately 20 percent of such tumors with approximately 350 to 400 new cases diagnosed each year in the United States. Approximately 70 percent of patients are diagnosed before the age of 20 years, and the peak incidence is between 5 and 9 years. The disease is rare after the fourth decade of life. The treatment of medulloblastoma in both children and adults, its prognosis, and the delayed complications in survivors will be reviewed here. The pathogenesis, clinical features, and diagnosis of medulloblastoma are discussed separately. PROGNOSTIC FACTORS — The most important factors affecting outcome are the extent of disease and the age of the patient at diagnosis. Molecular markers are being evaluated and may eventually permit further refinement in the choice of therapy. Extent of disease — The modified Chang criteria, based upon the extent of the primary medulloblastoma and associated metastases, are used to estimate prognosis, which is progressively worse in the presence of more advanced disease The spinal cord is a frequent site of spread, with involvement at presentation in 20 to 25 percent of patients. Contrast-enhanced magnetic resonance imaging (MRI) and CSF cytology are used to assess the craniospinal axis; this approach has replaced myelography. Staging is usually performed within one to two weeks after surgery. The importance of the presence of metastases was illustrated in a trial of 188 children with medulloblastoma. Patients without metastatic disease (M0) had a significantly higher five-year progression-free survival (PFS) (70 percent versus 57 percent with M1 and 40 percent with M2, M3, or M4 disease). Similar results were noted in a second series of 173 consecutive patients with medulloblastoma, 84 percent of whom were less than 15 years old. The five-year survival of those with M0 or M1 disease was 78 percent compared to 21 percent in patients with M2 or M3 involvement. Age at diagnosis — Young age at diagnosis has a negative impact on both long-term survival and quality of life in survivors. Children younger than age 5 years with medulloblastoma, and particularly those younger than 3, have a significantly poorer prognosis. This was illustrated by one series, in which the estimated five-year PFS was 32 percent between age 1.5 and 3 years versus 58 percent in those 3 and older Children diagnosed with medulloblastoma younger than age 3 years typically are treated with regimens that avoid or delay craniospinal radiation, because radiation is particularly damaging to the developing nervous system at this young age. Reduction or elimination of radiation in very young children may contribute to their overall poor prognosis. Although increased doses of RT can improve disease control, it is unlikely that this approach can be used to prolong survival in young children since the doses of RT are near the tolerated limit for normal brain and spinal cord. Further improvements in the outcome for young patients will require the development of effective new treatments that do not harm the developing nervous system. The natural history and response to treatment of medulloblastoma may differ in older as compared to younger children. In a retrospective review of 72 patients between the ages of 10 and 20 years at the time of diagnosis, five-year overall and event-free survival rates of 78 and 70 percent, respectively, were similar to those observed in younger children. However, relapses were often delayed with a mean time to relapse of 3.2 years. Furthermore, treatment in this older cohort was associated with severe ototoxicity, peripheral neuropathy, hematologic toxicity, and weight loss greater than 10 percent in 45, 71, 95, and 73 percent of patients, respectively. Molecular markers — For patients undergoing treatment of medulloblastoma, outcome appears to be influenced by the molecular features of the tumor. As an example, favorable prognosis medulloblastomas express high levels of the neurotrophin-3 receptor TrkC, low levels of myc oncogene transcription, and low expression of erbB-2 receptor. Analyzing the expression of a panel of genes using DNA microarray techniques (gene expression profiling [GEP]) may provide a more accurate reflection of prognosis than assessing single gene markers such as erbB-2 and may be a more powerful predictor of outcome than stage, histopathology, or age. Further study is necessary to establish the utility of GEP in clinical practice. PRINCIPLES OF TREATMENT — The appropriate treatment of patients with medulloblastoma includes the management of increased intracranial pressure and specific therapy directed against the tumor. Based upon results from multiple cooperative group trials, optimal treatment of medulloblastoma directed at the tumor includes a combination of maximal surgical resection, radiation therapy (RT) to both the tumor site and the craniospinal axis, and systemic chemotherapy. The application of this combined modality approach to different age groups is discussed below. Increased intracranial pressure — Children with medulloblastoma not uncommonly present with increased intracranial pressure due to obstructive hydrocephalus from compression of the fourth ventricle by the expanding tumor. Placement of a cerebrospinal fluid (CSF) shunt to relieve hydrocephalus is usually deferred until surgical resection since surgery alone is often sufficient to treat this problem. Local swelling from the tumor can contribute to symptoms of increased intracranial pressure. This vasogenic tumor edema is typically relieved by treatment with corticosteroids. The management of increased intracranial pressure is discussed separately. Surgery — Surgery is used to remove as much of the tumor as possible without causing serious neurologic sequelae (eg, persistent ataxia, cranial nerve deficits). Because of the potential for neurologic complications, total or radical resection is not always possible Patients who undergo a gross total or subtotal resection for nonmetastatic medulloblastoma have a better outcome than those who have a biopsy alone followed by RT. This was illustrated in two series, in which the five-year actuarial survival was higher in patients with complete or subtotal gross resection compared to biopsy alone (69 versus 40 percent, and 78 versus 43 percent. Posterior fossa local control rates are also higher in those who are managed with complete or subtotal resection (83 and 89 versus 27 percent with biopsy alone). Radiation therapy — Radiation therapy is an integral component of the initial management of patients with medulloblastoma, both to control residual posterior fossa disease and to treat any disease that has spread along the craniospinal axis. However, toxicity to the brain and spinal cord limits its usefulness. This is particularly true in very young children, for whom craniospinal radiation is avoided or delayed due to severe toxicity to the rapidly developing central nervous system. Technique — After surgical excision, medulloblastomas are treated with external beam RT therapy to the craniospinal axis, with an additional boost to the tumor site The rationale for additional radiation to the tumor bed is based upon the observation that 50 to 70 percent of recurrences occur in the posterior fossa. Although some investigators include the entire posterior fossa in the boost volume, posterior fossa failures are primarily in the tumor bed and are often associated with leptomeningeal failure. Local failure in the posterior fossa outside the tumor bed is rare as the solitary site of failure. Most protocols also include a stereotactic radiotherapeutic boost to residual tumor nodules if this can be done safely. Investigational approaches are examining the utility of proton beam and intensity-modulated RT Using conventional techniques, the brain and spine are treated with separate but overlapping RT fields. Placement of the junction between these two fields can have a significant impact on the cervical spinal cord and surrounding structures. Adjusting the site where these two fields overlap can minimize the dose of radiation to the spinal cord. However, sparing the spinal cord increases the dose to the thyroid gland, mandible, pharynx, and larynx. In developing children this may increase the risk of late hypothyroidism or mandibular hypoplasia. Newer techniques, such as intensity modulated RT, are being evaluated to limit radiation to normal tissues]. Complications — Although higher doses of RT are associated with better tumor control irradiation of the craniospinal axis in children is associated with a significant incidence of neurologic complications, including neurocognitive impairment. This issue is discussed separately. Whenever possible, RT is delayed for children younger than 3 years of age to permit further development of the central nervous system. For older children, the dose of RT is minimized. The potential negative impact from the combined effects of the tumor and treatment was illustrated by a study of 222 children irradiated for a brain tumor before the age of 4; only one-third of adult survivors were able to have full-time employment and a normal life-style In addition to its effects on neurocognitive development, craniospinal RT can cause decreased skeletal growth resulting in significantly decreased adult height. This effect appears to be mediated by decreased levels of growth hormone and may be partially avoided with decreased doses of radiation. The risk of other delayed effects of RT, including hypothyroidism, adrenal insufficiency, and hypogonadism, may also be minimized with lower doses of RT. Because of the risks of serious complications, the initial management of pediatric patients with medulloblastoma has utilized adjuvant chemotherapy with decreased doses of RT in average-risk children or substituted chemotherapy for RT in the initial management of infants and young children. Chemotherapy — Chemotherapy has an important role in the multimodality management of children with medulloblastoma in several settings:
CHILDREN — The combined modality approach to the treatment of children with medulloblastoma has evolved in a stepwise fashion. Radiation therapy was incorporated to reduce the rate of local recurrence in the surgical bed and along the craniospinal axis Subsequent multicenter randomized trials integrated chemotherapy along with radiation therapy following surgical resection. Although randomized trials did not demonstrate a statistically significant benefit to overall survival, subset analyses suggested that chemotherapy did improve survival in some patients. A subsequent trial demonstrated an improvement in event-free survival, although the difference in overall survival was not statistically significant. Most of these trials included children with varying levels of risk for recurrence. Furthermore, advances in imaging, surgery, and radiotherapy techniques limit comparison of current results with those of earlier trials.
Average-risk children — The current
approach to children age
Children's Oncology Group trial — In 2006, the Children's Oncology Group (COG) published results from a large trial, which has defined the current standard of care for children with average-risk medulloblastoma and illustrates the outcomes using a combined modality approach that includes surgery, RT, and chemotherapy. This phase III trial included 421 children who had undergone complete or nearly complete surgical resection (residual disease <1.5 cm2 on postoperative imaging). Patients were required to have no evidence of disseminated disease on MRI and a negative cytologic examination of cerebrospinal fluid. Following surgery, children were treated with 23.4 Gy of craniospinal RT, followed by a posterior fossa boost of 32.4 Gy, for a total RT dose of 55.8 Gy to the tumor bed. Children received weekly vincristine during RT. After completion of RT, patients were randomly assigned to eight cycles of chemotherapy with one of two regimens (vincristine, cisplatin, plus CCNU, or vincristine, cisplatin, plus cyclophosphamide). For the entire cohort of 379 evaluable children, the five-year event-free and overall survival rates were 81 and 86 percent, respectively, at a median follow-up of over five years. There were no significant differences between the two chemotherapy regimens. In addition, in 15 patients who were treated according to the protocol but were deemed to be ineligible at subsequent review based upon the presence of excess residual disease or disseminated disease at randomization, the five-year event-free survival was only 36 percent. The event-free survival in average-risk patients in this trial incorporating adjuvant chemotherapy is superior to that previously seen in trials using RT alone, even when a higher dose of RT is used. The results are also at least as good as those seen in trials using more intensive chemotherapy regimens. Toxicity associated with treatment in these children included the following:
Future directions — Because of the neurologic toxicity associated with this dose of RT in other studies, the ongoing medulloblastoma trial of the COG is randomly assigning children to craniospinal RT at a dose of either 18 Gy or 23.4 Gy, with a boost to give a total dose of 54 Gy to the tumor bed. This trial will look at both event-free survival and toxicity as end points. ADULTS — The optimal treatment of adults with medulloblastoma is unclear, and there are no large series or randomized trials. Therapy is patterned after that in children, with RT used for low-risk disease and combined RT and chemotherapy for high-risk patients. One prospective study enrolled 36 patients 18 years of age into a treatment protocol, stratifying patients according to stage
The five-year PFS was significantly better in patients with low-risk disease (76 versus 61 percent), and was similar to the outcomes observed in children. The reported five- and ten-year survival rates in adults are 50 to 60 and 40 to 50 percent, respectively. However, late relapses, lateral tumor location, and desmoplastic histologic features are more common in adults SUMMARY AND RECOMMENDATIONS — Combined modality therapy, including surgery, craniospinal radiation therapy (RT), and chemotherapy is the standard of care for both children and adults. Patients should be treated in clinical trials whenever possible.
- The protocol used in the Children's Oncology Group medulloblastoma trial is the current standard. After gross total resection of the tumor, patients are treated with 23.4 Gy to the craniospinal axis with a posterior fossa boost to a total dose of 55.8 Gy. This is followed by eight cycles of chemotherapy with cisplatin, vincristine, and either cyclophosphamide or CCNU.
- For adults with low-risk disease (T1-3a,M0, and no residual disease after surgery) we suggest treatment with craniospinal RT to 36 Gy, supplemented by a local tumor boost for a total tumor dose of 55.8 Gy - For adults with high-risk disease (T3b-4, any M+, or postoperative residual tumor,we suggest chemotherapy followed by craniospinal RT (). If M1, M2, or M3 disease was present at diagnosis, maintenance chemotherapy is indicated. |
3
years of age with medulloblastoma that is amenable to total or near-total
resection includes surgery, followed by craniospinal irradiation, and then
adjuvant chemotherapy. The results of this approach in this patient subset
are illustrated by the Children's Oncology Group trial.
3
years of age with medulloblastoma are at high risk of severe neurologic
impairment if their initial treatment includes craniospinal RT. We
recommend that treatment in this age group be done on approved protocols
that utilize combination chemotherapy to either delay or obviate the
need for craniospinal RT