About Bioline  All Journals  Testimonials  Membership  News  Donations

Indian Journal of Cancer
Medknow Publications on behalf of Indian Cancer Society
ISSN: 0019-509X EISSN: 1998-4774
Vol. 39, Num. 3, 2002, pp. 97-105

Indian Journal of Cancer, Vol. 39, No. 3, (July - September 2002), pp. 97-105

Medulloblastomas: Clinical profile, treatment techniques and outcome - An Institutional experience

Punita Lal, Yoodhvir Singh Nagar, Shalini Kumar, Shalim Singh, K. J. Maria Das, S. Lakshmi Narayan, Raj Kumar,* V. K. Jain, Sundar Ayyagari

Department of Radiotherapy, and *Department of Neurosurgery, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow 226 014, India.

Code Number: cn02003


The present series reports an audit on the patterns of presentation, radiation treatment techniques, failure pattern and outcome in the 36 patients treated at a single institution. Patients were accrued between October 1991 and September 1999. They underwent total or subtotal resection along with craniospinal irradiation. The dose to the cranium ranged from 30 to 43Gy (median- 36Gy), to the spine from 20 - 36Gy (median- 36Gy) and the posterior fossa boost, which was delivered in 32 cases, ranged from 14 to 24Gy (median -18Gy). Simulator film evaluation was carried out at the time of analysis based on the French Medulloblastoma Group guidelines, which revealed a significant under-dosage in the region of posterior fossa and cribriform plate in 27% and 19% respectively. Adjuvant chemotherapy was administered in 7 patients. Salvage treatment in the form of chemotherapy (5 cases) and re-irradiation (2 cases) were attempted but were ineffective. The overall survival (OS) and progression free survival (PFS) rates were 54% and 40% respectively, with the median being 75 and 29 months respectively. Amongst all the prognostic factors considered, on univariate analysis, duration of symptoms was significant for PFS with a trend towards significance for OS, while extent of debulking had a trend towards significance for PFS. No factor emerged significant on multivariate analysis.

Key Words : Medulloblastoma, Radiotherapy, Chemotherapy, Surgery, Patterns of failure.


Medulloblastoma (MB) comprises 20% of all childhood intracranial neoplasms. Although commoner in childhood, infrequently they tend to occur in adults as well.1 In view of the rarity of this disease, clinical experience in most institutions would be limited and prognostic factors known with less certainty. This report is a single institution audit of patients with medulloblastoma referred over a decade to the department of radiotherapy for post-surgical adjuvant therapy. The purpose is to highlight the clinical profile, treatment techniques, and patterns of failure besides attempting to elicit prognostic factors likely to influence treatment strategies and outcomes.


Between October 1991 and September 1999, 36 patients of MB were referred to the department of Radiotherapy from the department of Neurosurgery, for further management. Information was recorded from case charts about demographic characteristics, clinical symptoms, radiographic findings based on contrast enhanced computed tomography (CECT) or magnetic resonance imaging (MRI) scans. The extent of resection was based on surgeons operative notes while the treatment offered and response and patterns of recurrence were noted. All the histopathology reports originated from the department of pathology at this institute, and were not reviewed again for the purpose of this analysis.

Pretreatment assessment

The usual preoperative imaging for the patients consisted of a CECT before 1995. After this period, MRI was increasingly used as the imaging modality of choice. The spinal cord was not imaged routinely in these patients. Cranio-spinal fluid (CSF) was routinely sampled after 1997 when the patient reported for radiotherapy (RT), which was usually 2 weeks or more following surgery. In view of these limitations, we could not ascribe Chang's staging2 retrospectively to these patients, as there was incomplete information about the disease status in spinal cord.

Surgical resection

The surgeon usually attempted maximal surgical resection compatible with preservation of neurological function, and so surgical procedures ranged from partial to total removal of the tumor. Shunt procedures were performed when indicated. The volumetric extent of resection was based on operative notes mainly, as an imaging study of the primary site was not uniformly performed for all patients in the immediate postoperative phase.

Radiotherapy protocol

All patients were offered cranio-spinal RT. The patients were simulated in the prone position with an immobilization cast for the cranium only. `Body immobilization frames' were not used. Telecobalt with a source to axis distance of 80cm was used. The entire cranium was treated with a `German helmet technique' to encompass all the intracranial contents with custom blocking fabricated for each patient. The plan usually consisted of the entire neuraxis dose of 30-36Gy at 150 to 180cGy per fraction, while the posterior fossa was to receive a dose of 50-55Gy at 150­200 cGy per fraction. The usual technique involved moving the cranio-spinal junction at every 12Gy with the cranial and spinal fields treated daily. The spinal cord was treated down to the S2­S3 junction with usually 5-6cm wide fields in the thoraco-lumbar region and a spade like expansion in the sacral region. No compensators were used. The depth of prescription for the cranial contents was at mid depth along the central axis while that for the spinal region was usually at the posterior vertebral body. Blood counts were routinely performed twice a week in all patients, and RT omitted if the total leukocyte count fell below 3000/mm3. Steroids and anti emetics were instituted as and when indicated. Adjuvant chemotherapy was administered in only a few patients based on the physician perception of high-risk patients i.e. CSF involvement/spinal seedling These usually consisted of nitrosoureas or combination of agents.

Radiotherapy technique

An audit of simulation films of the whole brain (WB) and of the posterior fossa (margins) was performed to assess the correctness of technique and evaluate the potential influence of adequacy of margins on relapse rates. Observers who were aware of the treatment outcome did this evaluation, which was therefore not blinded. Fields were considered acceptable if portal limits as drawn on simulation films surrounded the brain target with at least 5 mm margin for the orbit (10 mm for other sites) when using telecobalt. A marginal miss was scored when the margin of error was 3-5 mm for the orbits (5 -10mm for other sites) and a major miss if the margin was <3 mm for the orbits (<5 mm for other sites). Posterior fossa fields were considered acceptable only if the anterior limit of the portals were at the posterior clinoids, the inferior margin at C1 / C2 junction, superior margin 1 cm above the point mid way between foramen magnum and the vertex while the posterior margin behind the inion. No port films were taken during treatment.

Response assessment and follow up

Response evaluation criteria were based on change in objective clinical signs and subjective improvement. Failure was specified by site whenever available and elaborated as far as supra­tentorium was concerned. (Figures 1 & 2). Radiological assessment of response to treatment could be assessed only in those patients where such studies had been performed and the films available for review. A complete response (CR) was defined clinically based on symptom improvement and radiologically whenever there was complete resolution of tumor as documented on CECT or MRI scans. The two methods have been reported separately.

Survival analysis

Time to tumor progression (TTP) and overall survival (OS) were estimated by the Kaplan-Meier method and computed from the date of surgery. Disease or symptom progression (in the absence of documentation by imaging) was considered as an event for TTP while death due to any cause was considered an event for both TTP and OS. The log-rank test was used for univariate comparison of prognostic variables. All tests were two-sided, and p-values of < 0.05 were reported significant.

Multivariate analysis was done using Cox's proportional hazards model. Quality of life issues and neurocognitive sequalae were not addressed in this audit.


Patient and disease characteristics (Table 1)

The median age of presentation of these 36 patients was 10 years (range 2 - 34). Eight (22%) of these were 15 years or older. Male preponderance was observed in the ratio of 2.6:1. The median duration of symptoms from onset to diagnosis was 5 months (range 1 to 60), headache (81%) and vomiting (72%) being the commonest presentation. Apart from the posterior cranial fossa lesion, 4 patients (11%) showed evidence of ventricular involvement. Most tumors appeared to be either mixed attenuating (33%) or showed increased attenuation on CECT imaging. Cytological assessment of cerebrospinal fluid for malignant cells was done in 11 cases, of which 3 were reported positive.

Interventions (Table 2)

Fifty three percent (19/36) had a shunt placed prior to definitive surgery. Surgery was carried out in the form of gross total/ subtotal tumor excision (considered as >50% removal for the purpose of this study) or lesser resections in the form of tumor debulking (considered as <50% removal). Based on the available operative details, it was inferred that > 50% resection was performed in 89% (32/36). Radiotherapy was commenced within 4 weeks in about 39% of these patients. The delay in other patients could be ascribed to prolonged convalescence (17%), postoperative complications (19%), non-compliance of instructions by the patients (17%) or no cause could be ascribed (8%).

Thirty-two out of 36 patients completed the planned treatment of cramospinal irradiation. The spine was irradiated to the planned dose of 30 Gy in 28% and 36Gy in 64%. All but 11% (4/36) patients received 50 Gy or higher to the posterior fossa, with 64% receiving 54Gy or higher. The whole brain and spine received a median dose per fraction of 1.5Gy, while it was higher at 1.95Gy during the posterior fossa boost. Reasons for non-compliance in the 4 patients could not be elicited from the case charts but were presumably treatment toxicity related.

Radiation was tolerated reasonably well with acute reversible toxicity seen in 75% (27/36) (Table 3). Treatment interruptions were necessary in half the patients for reasons of acute reactions (25%), infection (3%), social reasons and holidays (8%) and no cause ascribed (14%). The mean duration of delay in these 18 patients (defined as time beyond 50 days) was 10.78 days (S.D. 10.08; range 2-37).

Chemotherapy was given in 7 patients based on the physician's perception of high risk for relapse following CSI alone and consisted of single agent CCNU (3 patients) or PCV (4 patients).

Response and patterns of failure

Responses were assessed in the follow up period, usually a month after completion of all RT (Table 4). Based on clinical criteria, 59% (23/36) had marked improvement in symptoms and/or signs. Radiological assessment could be performed in 16 patients in whom a complete set of pre and postoperative films could be reviewed. Fourteen of these patients had complete radiological resolution of disease of which 11 (79%) had marked improvement clinically.

Patterns of failure could be documented in 8 patients. Three had local failure in the posterior fossa. Two relapsed in the subarachnoid space, 1 had extra neuraxial metastasis in the lung, and while the remaining 2 had basi-frontal metastasis. None of these patients had had a radiologically documented complete resolution of tumor.

An attempt was made to see if there was any association between adequacy of radiation field placements and intra-cranial relapse. These were based on the guidelines adapted from the French Medulloblastoma Group4 and Swiss study.5 A perusal of table 5 shows that although the incidence of major deviations in the region of the cribriform plate was of the order of 19%, in only one of the two recurrences documented in the basi-frontal area occurred in this group. It is conceivable how ever, that the other patients who relapsed and died in the follow up period also had recurrences in this area, but there is no documentation to this effect as films were not available. Similarly all the 3 who failed in the posterior fossa had documented tighter volumes (1 or more inadequate margins) during the boost margins.

The median progression free interval in the present group was 29 months (range 0 - 114 months) and median overall survival at 75 months (range 4 - 114 months). The progression free survival (PFS) and overall survival (OS) at 5 years was 40% and 54% respectively (Figures 3 & 4). The prognostic factors considered in this analysis were age, gender, duration of symptoms, shunt placement, pre operative KPS, extent of surgery, post operative KPS, gap between surgery and radiotherapy, RT duration and RT dose (Table 6). Amongst these, on univariate analysis, duration of symptoms was significant for PFS with a trend towards significance for OS, while extent of debulking had a trend towards significance for PFS. No factor emerged significant on multivariate analysis.


The primary purpose of this audit was to identify prognostic factors which could affect progression free survival and over all survival. This audit has shown that probability of OS is about 54% at 5 years, quite similar to the 58% reported on 150 patients from a very contemporary series in which 16% of the population of patients consisted of adults.6,9 Medulloblastoma is a tumor of childhood, but is known to occur in adults as well.7 Medulloblastoma occurring in children less than 3 years bears an unfavorable prognosis.8 However, most series have shown that apart from the very young children, age is not regarded as a significant prognostic factor, the outcome being more dependent upon disease related and treatment related factors. Age did not emerge as a factor of significance in the present series.

In up to 40% of the patients, MBs may extend in to the adjacent brain stem at presentation. These tumors are characterized by subarachnoid seeding of the leptomeninges, occurring in 30-50% cases at the time of presentation.5 In the present series, CSF was evaluated for the presence of malignant cells in only 30%(11,36) of these patients. Three had evidence of dissemination of cells in the CSF. Spinal imaging was not performed in any patient. Therefore, it has not been possible to determine the cases with dissemination or determine the prognostic significance of the same.

There is increasing agreement in the literature about the importance of surgical resection in this disease. Although, resectability rates vary from 40-70%, the more important variable appears to be the degree of post-operative residual. It is recommended that in order to determine the degree of post-operative residual, plain and contrast enhanced CT scans or plain / gadolinium enhanced MR scans of the posterior fossa be performed within 3 days of surgery, before post operative changes mimic residual tumor and potentially flaw the measurement. One of the conclusions of the Children's Cancer Group 921 trial addressing the relative efficacy of two different chemotherapy regimens along with standard RT doses was the adverse effect of a residual of > 1.5cm2 in patients more than 3 years of age and with MO disease.10 We estimated the degree of resection (greater or lesser than 50%) based on the surgeons' perception of removal rather than the objective method of imaging. The commonest reason for a repeat imaging in the immediate postoperative period was to determine a correctable cause of deterioration in the postoperative period and not to document residual disease. Not withstanding these reservations and of a grossly dissimilar sample size, we documented a trend (p=0.06) of an improved PFS for patients with > 50% resection on univariate analysis.

For the past seven decades tumor resection surgery followed by craniospinal irradiation has been the established mode of treatment for these tumors. A contemporary estimate of the outcome following surgery and RT was provided by the first POG-CCG low-stage medulloblastoma trial. Patients with localized disease following total or near total resection and conventional RT achieved a 70% event free survival at 5 years.11 We followed a similar policy of 50-54 Gy to the posterior fossa and 30-36 Gy to the cranio-spinal axis. All but 4 patients received 50Gy or higher to the posterior fossa. Three of them relapsed in the posterior fossa while the fate of the last one was not known. All but 3 patients received at least 30Gy to the spine. These interventions were generally well tolerated with grade 3 neutropenia seen in only 11 % patients.

The French Medulloblastoma Group 4 published their results of the contribution of radiotherapy technique to tumor relapse in a multicentric prospective study. The whole brain RT was recommended by the "German Helmet" technique to a dose of 27 Gy to the supratentorial compartment, 36Gy to neuraxis at 180cGy dose per fraction. And, a boost up to 50-55 Gy was advised to the tumor bearing area with margin or to the entire posterior fossa. The impression was that underdosage in the region of the cribriform plate was the cause of frontal relapses seen in these cases. One of 6 with a major deviation and 1/18 with a minor deviation in the cribriform region had a subfrontal relapse in the present series. Retrospective assessment of simulator films however, required further correlation with port films for repositional accuracy documentation at the time of treatment.

With regard to the posterior fossa, the French Medulloblastoma Group felt that treating the gross tumor with a 2 cm margin was sufficient, and did not increase the recurrence rate. In this study, no correlation between adequacy of portals and recurrences in the posterior fossa could be documented. Use of photons or electrons did not influence the failure pattern, in the spine. Spinal dose in the range of 30-36Gy was considered sufficient, but it was suggested that homogeneity must be attempted in this region. In this study, only telecobalt was used to treat the spine. Miralbell et al in 1997 retrospectively analysed their results in 86 patients. They also observed that majority of post treatment failures occurred in the posterior fossa, which was closely followed by supratentorial failure. Cribriform plate relapse, due to excessive eye shield was the commonest site. It was studied in this sitewise analysis that field placement correlated significantly with the supratentorial failure free survival.5 No such correlation could be drawn from the present analysis, due to smaller numbers.

In adjuvant setting, the role of chemotherapy is emerging both in high as well as low grade tumors. In the high grade MBs addition of chemotherapy has yielded better survival, whereas, in the low-grade variety, its introduction has managed to reduce the radiation dose to the neuraxis, thereby decreasing the neuropsychological sequelae.11,12 Evans et al. in a phase III trial administered CCNU, prednisone and vincristine in an adjuvant fashion. The first two agents were administered since each had a different mechanism of action, and had individually shown responses in recurrent medulloblastomas. Vincristine, although not crossing the blood brain barrier was known to be active against tumor which, it was hypothesized, had no such barrier. The lipid solubility of CCNU has made it an attractive antineoplastic agent for brain tumors.13

Although this was a negative trial for the entire cohort, chemotherapy appeared to be most useful for the subset of T3, T4 and M1-3 subset of patients. Packer et al. treated the `poor risk' medulloblastoma group with radiation plus adjuvant chemotherapy i.e. CCNU, vincristine and cisplatin. They observed a significant difference in survival 35 % versus 87 % (p<0.001) in favour of chemotherapy although realized the oto-toxic potential of cisplatin.14 The use of chemotherapy in this cohort of patients was largely based on individual physicians perceptions of patients with poor risk factors. No apparent benefit was observed, but no inference can be derived from this study.

During follow up, neuro cognitive and neuro psychological assessment were not carried out in these patients, but subjectively, there was no significant alteration in the quality of life and academic/mental performance of these patients, once they were free of disease. Similarly, neurologically, definite subjective improvement was observed in all the surviving patients. Amongst the prognostic factors that were considered, what was unusual was the observation that patients with symptoms of longer duration than 4 months had significantly better PFS and a trend towards improvement of the OS as well. This may be that longer duration of symptoms could represent a biologically more indolent disease. However, this variable did not emerge as being of independent prognostic significance of multivariate analysis. We did not find KPS to be of prognostic significance as suggested by Carrie et al.15 Perhaps, it is the small sample size that has plagued this study and therefore, we have been unable to elicit factors of prognostic significance To summarize, a perusal of literature and our own experience suggests that an aggressive surgical approach followed by optimal neuraxis radiation continues to remain the mainstay of treatment. Since medulloblastoma is one of the tumor where quality of treatment has been proven to affect the disease control, radiotherapy should be meticulously planned at centers where facilities for immobilization, simulation and conformal shielding blocks are available. Chemotherapy may supplement radiotherapy in the high-risk groups and has been successful in reducing the radiation doses in low risk groups.16


  1. Hai Meder C, Song PY. Medulloblastoma: differences in adults and children. Int J Radiat Oncol Biol Phys 1995;32:1255-7.
  2. Chang CH, Housepian EM, Herbert C Jr. An operative staging system and a megavoltage radiotherapeutic technique for cerebellar medulloblastomas. Radiology 1969;93:1351-9.
  3. Miller AB. Reporting results of cancer treatment. Cancer 1981;47:207-14.
  4. Carrie C, Hoffstetter S, Gomez F, Monvho V, Doz F, Alapetite C, et al. Impact of targeting deviations on outcome in medulloblastoma: Study of the French Society of Pediatric Oncology (SFOP). Int J Radiat Oncol Biol Phys 1999;45:435-9.
  5. Miralbell R, Bleher A, Huguenin P, Ries G, Kann R, Mirimanioff RO, et al. Pediatric medulloblastoma: radiation treatment technique and patterns of failure. Int J Radiat Oncol Biol Phys 1997;37:523-9.
  6. Jenkin D, Shabanath MA, Shail EA, et al. Prognostic factors in Medulloblastoma. Int J Radiat Oncol Biol Phys 2000;47:573-84.
  7. Sharma BS, Kak VK, Mittal RS, Banerjee S, Ayyagari S. Medulloblastoma in adults - clinical observations and results of treatment. Ind J Cancer 1989;26:156-63.
  8. Packer RJ, Goldwein J, Nicholson HS, Vezina G, Allen JC, Ris MD, et al. Treatment of children with medulloblastomas with reduced dose of cranial radiation therapy and adjuvant chemotherapy: A Children's Cancer Group Study. J Clin Oncol 1999;17:2127-36.
  9. Giangspero F, Perilongo G, Fondelli MP, Brisigotti M, Carollo C, Burnelli R, et al. Medulloblastoma with extensive nodularity: a variant with favorable prognosis. J Neurosurg, 1999;91:971-7.
  10. Zeltzer PM, Boyett JM, Finlay JL, Albright L, Rorke LB, Milstein JM, et al. Metastasis Stage, Adjuvant treatment, and residual tumor are prognostic factors for medulloblastoma in children: conclusions from the Children's Cancer Group 921 randomized phase III study. J Clin Oncol 1999;17:832-45.
  11. Deutsch M, Thomas PR, Krischer J, Boyett JM, Albright L, Aronin P, et al. Results of a prospective randomized trial comparing standard dose neuraxis radiation with reduced neuraxis radiation in patients with low stage medulloblastoma: a combined Children's Cancer Group - A Pediatric Oncology Group Study. Pediatr Neurosurg 1996;24:167-76.
  12. Levin VA, Rodriguez LA, Edwards MSB, Wara W, Lui HC, Fulton D, et al. treatment of medulloblastoma with procarbazine, hydroxyurea, and reduced radiation dose to whole brain and spine. J Neurosurg 1988;68:383-7.
  13. Evans AE, Jenkin DT, Sposio R, Ortega JA, Wilson CB, Wara W, et al. The treatment of medulloblastoma. Results of a prospective randomized trial of radiation therapy without CCNU, vincristine, and prednisone. J Neurosurg 1992;72:572-82.
  14. Packer RJ, Sutton LN, Goldwein JW, Perilongo G, Bunin G, Ryan J, et al. Improved survival with the use of adjuvant chemotherapy in the treatment of medulloblastoma. J Neurosurg 1991;74:433-40.
  15. Carrie C, Lasset C, Alapetite C, Haie-Mader C, Hoffstetter S, Damaille MC, et al. Multivariate analysis of prognostic factors in adult patients with medulloblastoma. Retrospective study of 156 patients. Cancer 1994;74:2352-60.
  16. Kortman RD, Kuhl J, Timmermann B, et al. Postoperative neoadjuvant chemotherapy before radiotherapy as compared to immediate radiotherapy followed by maintenance chemotherapy in the treatment of medulloblastoma in childhood: Results of the German prospective randomized trial HIT' 91. Int J Radiat Oncol Biol Phys 1999;46: 269-79.

Copyright 2002 - Indian Journal of Cancer.

The following images related to this document are available:

Photo images

[cn02003t4.jpg] [cn02003f4.jpg] [cn02003f3.jpg] [cn02003t3.jpg] [cn02003t6.jpg] [cn02003f1.jpg] [cn02003f2.jpg] [cn02003t1.jpg] [cn02003t5.jpg] [cn02003t2.jpg]
Home Faq Resources Email Bioline
© Bioline International, 1989 - 2022, Site last up-dated on 11-May-2022.
Site created and maintained by the Reference Center on Environmental Information, CRIA, Brazil
System hosted by the Internet Data Center of Rede Nacional de Ensino e Pesquisa, RNP, Brazil