Radiotherapeutic Management of Medulloblastoma

June 1, 1997

Dr. Paulino provides a concise yet complete review of the radiotherapeutic management of patients with medulloblastoma. Radiotherapy treatment planning for medulloblastoma is complex, requires considerable attention to detail, and remains the subject of debate and clinical research. Clearly, this is an area of neuro-oncology in which multidisciplinary research has played a significant role in improving survival for children and young adults with this disease.

Dr. Paulino provides a concise yet complete review of the radiotherapeuticmanagement of patients with medulloblastoma. Radiotherapy treatment planningfor medulloblastoma is complex, requires considerable attention to detail,and remains the subject of debate and clinical research. Clearly, thisis an area of neuro-oncology in which multidisciplinary research has playeda significant role in improving survival for children and young adultswith this disease.

Ironically, perhaps, research on the radiotherapeutic management ofmedulloblastoma has resulted in a shift from less treatment (treatmentof the tumor volume only prior to 1950s), to more treatment (treatmentof the neuraxis), to attempts to treat less again (reduction in dose tothe neuraxis, deferral of radiation with the use of chemotherapy). Evenin terms of chemotherapy, there is now a strong bias among many investigatorsto use adjuvant chemotherapy in all stages of medulloblastoma, whereasin the past its use was restricted to poor-risk patients. Indeed, the definitionof "risk" has undergone significant change over the last decade,depending less on Chang stage classification for local tumor burden andmore on the extent of postoperative residual disease as measured by MRI.

All of these changes have come about because of careful observationand analysis of clinical outcomes, primarily because of clinical researchfrom the pediatric cooperative groups in the United States and Europe.Two areas of research are particularly important, and in many ways linked:(1) attempts to reduce toxicity by lowering the craniospinal dose of radiationand (2) efforts to improve progression-free survival in low-risk patientsby adding adjuvant chemotherapy.

Attempts to Reduce Radiation Dose

Dr. Paulino thoroughly discusses the attempts that have been made toreduce radiation dose to the brain and spine by minimizing actual totaldose delivered or altering the fractionation scheme to potentially reducelate toxicity, as well as by the addition of chemotherapy. These approacheshave had mixed results.

Various reports from single institutions using lower-dose craniospinalradiotherapy (< 30 Gy) have been published.[1,2] Single-fraction doseschedules have been used most frequently, often reducing the spinal doseto 24 Gy. In patients with good risk factors, progression-free and overallsurvival rates appear to be comparable to those seen historically usinghigher-dose treatment (36 Gy). These studies have had sufficient heterogeneitywith regard to patient populations and actual treatment used, however.Thus, the results could have been biased due to patient selection factors.

A phase III study, CCG 923/POG 8631, eventually was initiated in patientswith low-risk features. This study randomized patients to standard-doseor lower-dose radiation. Initial results seemed to suggest that the lower-doseradiation arm would result in higher rates of exoprimary failure, and thestudy was suspended prior to full accrual. A more recent analysis of thisstudy cast doubt on that observation, and it is possible that the studyquestion in this now closed trial will not be answered as a consequence.[3]

A second intergroup study, CCG 9014/POG 9331, randomized patients withgood-risk features to standard-dose radiation or lower-dose craniospinalradiation with adjuvant chemotherapy. Unfortunately, this study also closedearly, this time because of insufficient accrual, suggesting a significantbias toward these approaches among investigators as well as patients. Thus,the question of whether one can lower the dose of craniospinal radiotherapyhas still not been answered.

Additional attempts to reduce craniospinal dose have included the useof hyperfractionation techniques and the addition of chemotherapy to radiationin lower-risk groups similar to the patients in the CCG 9014/POG 9331 study.For the most part, these trials have been uncontrolled.


Our group and others began to use hyperfractionation schedules withthe hope of safely increasing dose to the local tumor while decreasingdose to the brain and spine. Lower doses per fraction were used in an attemptto minimize late effects in nonproliferative neural tissue, and more frequentdosing was employed to potentiate cytotoxic effects in acute-reacting tissue,such as tumor. We learned early on that such an approach was inadequatein terms of preventing exoprimary relapse when 24 Gy was given to the spineand brain in patients without obvious disease in those areas. Clearly,24 Gy was "biologically" inappropriate to control tumor in areasof the brain and spine at risk of relapse. However, in selected patients,the delivery of 72 Gy to the primary tumor site in the posterior fossawas found to be effective in controlling local disease, clinically tolerable,and without the long-term risk of radiation necrosis.[4]

In CCG 9931, higher doses to the spine and brain are now being testedfollowing intensive preradiation chemotherapy in patients with high-riskfeatures. It may be possible to improve local tumor control rates, a significantclinical problem, by increasing dose to the primary and metastatic tumorsites through the use of hyperfractionation techniques. Many questionsstill remain about this approach, including the most basic ones, such asefficacy relative to standard single-fractionation schemes, appropriatedose, and late effects. Thus, hyperfractionation must still be consideredinvestigational.

The use of adjuvant chemotherapy has traditionally been reserved forpatients with high-risk features, such as those with residual disease followingsurgery or tumor spread beyond the primary site. Recent reports stronglysuggest that, in some high-risk patients, the addition of chemotherapymay result in equal or better 5-year progression-free survival rates thanthose observed in patients with low-risk factors who are treated with radiationalone.[5] These reports, most notably those using the combination of lomustine(CeeNu), cisplatin (Platinol), and vincristine following radiation, arecompelling, and have resulted in a new study being conducted by the CCGand POG (protocol A9961). This study will randomize patients with low-riskfeatures to one of two chemotherapy regimens following craniospinal radiation(2,340 cGy to the brain and spine and 5,580 cGy to the primary tumor site).This study is being conducted despite a number of previous controlled phaseIII studies showing no survival benefit from the addition of chemotherapyto radiation in low-stage patients.[6,7]

A Different Chemotherapy Regimen

The chemotherapy that will be used in the new study differs from theregimens used in these older studies. In the new study, cisplatin willbe added to lomustine and vincristine and will be compared to cyclophosphamide,cisplatin, and vincristine. The hope is that either of these chemotherapyregimens not only will "substitute" for the reduced craniospinalradiation dose, lessening the risk of exoprimary relapse, but also willresult in higher progression-free survival rates than have been seen previouslywith standard-dose radiation alone. All of these results need to be accompaniedwith less neurocognitive, endocrine, and spinal toxicity than has beenseen with higher-dose radiation. If this study accrues a sufficient numberof patients to keep it open for full enrollment, the answers to the studyquestions may be only a few years in coming. Thus, in 1997, low-risk patientsenrolled in this study will all receive chemotherapy.

It is also important to keep in mind that although 5-year progression-freesurvival rates of patients with medulloblastoma appear to be improving,10-year results are important as well. This is true in terms of actualprogression-free survival, as well as overall "quality" survivaland the potential risk of second malignancies in these children with low-riskfeatures who are treated with both radiation and alklyator-based chemotherapy.


The possibility of curing medulloblastoma does exist, particularly inpatients with low-risk features. Dr. Paulino's review is important andtimely, and should encourage physicians who treat these patients to continueto ask questions and to participate in clinical trials that attempt toanswer them.


1. Goldwein JW, Radcliffe J, Johnson J, et al: Updated results of apilot study of low dose craniospinal irradiation plus chemotherapy forchildren under five with cerebellar primitive neuroectodermal tumors (medulloblastoma).Int J Radiat Oncol Biol Phys 34:899-904, 1996.

2. Deutsch M, Thomas PRM, Krischer J, et al: Results of a prospectivetrial comparing standard dose neuraxis irradiation (3600 cGy/20) with reducedneuraxis irradiation (2340 cGy/13) in patients with low-stage medulloblastoma:A combined Children's Cancer Group-Pediatric Oncology Group study. PediatrNeurosurg 24:167-176, 1996.

3. Halberg FE, Wara WM, Fippin LF, et al: Low-dose craniospinal radiationtherapy for medulloblastoma. Int J Radiat Oncol Biol Phys 20:651-654, 1991.

4. Prados MD, Wara WM, Edwards MSB, et al: Hyperfractionated craniospinalradiation therapy for primitive neuroectodermal tumors: Early results ofa pilot study. Int J Radiat Oncol Biol Phys 28:431-438, 1993.

5. Packer RJ, Sutton LN, Elterman R, et al: Outcome for children withmedullablastoma treated with radiation and cisplatin, CCNU, and vin-cristinechemotherapy. J Neurosurg 81:690-698, 1994.

6. Tait DM, Thornton-Jones H, Bloom HJG, et al: Adjuvant chemotherapyfor medulloblastoma: The first multi-centre control trial of the InternationalSociety of Paediatric Oncology (SIOP). Eur J Cancer 26:464-469, 1990.

7. Evans AC, Jenkin RDT, Sposto R, et al: The treatment of medulloblastoma.Results of a prospective randomized trial of radiation therapy with orwithout CNU, vincristine, and prednisone. J Neurosurg 72:572-582, 1990.