Standard-Risk Medulloblastoma Treated by Adjuvant Chemotherapy Followed by Reduced-Dose Craniospinal Radiation Therapy: A French Society of
http://www.100md.com
《临床肿瘤学》
the Institut Curie, Paris
Institut Gustave Roussy, Villejuif
Centre Hospitalo–Universitaire (CHU) La Timone, Marseille
Centre Leon Berard, Lyon
CHU Rennes, CHU Nancy
CHU Grenoble
CHU Strasbourg
Centre Oscar Lambret, Lille
CHU Lille
Unite de l'Institut National de la Sante et de la Recherche Medicale U537, Le Kremlin Bicêtre, France
Hpital Reine Fabiola, Brussels, Belgium
ABSTRACT
OBJECTIVE: The primary objective of this study was to decrease the late effects of prophylactic radiation without reducing survival in standard-risk childhood medulloblastoma.
PATIENTS AND METHODS: Inclusion criteria were as follows: children between the ages of 3 and 18 years with total or subtotal tumor resection, no metastasis, and negative postoperative lumbar puncture CSF cytology. Two courses of eight drugs in 1 day followed by two courses of etoposide plus carboplatin (500 and 800 mg/m2 per course, respectively) were administered after surgery. Radiation therapy had to begin 90 days after surgery. Delivered doses were 55 Gy to the posterior fossa and 25 Gy to the brain and spinal canal.
RESULTS: Between November 1991 and June 1998, 136 patients (median age, 8 years; median follow-up, 6.5 years) were included. The overall survival rate and 5-year recurrence-free survival rate were 73.8% ± 7.6% and 64.8% ± 8.1%, respectively. Radiologic review showed that 4% of patients were wrongly included. Review of radiotherapy technical files demonstrated a correlation between the presence of a major protocol deviation and treatment failure. The 5-year recurrence-free survival rate of patients included in this study with all optimal quality controls of histology, radiology, and radiotherapy was 71.8% ± 10.5%. In terms of sequelae, 31% of patients required growth hormone replacement therapy and 25% required special schooling.
CONCLUSION: Reduced-dose craniospinal radiation therapy can be proposed in standard-risk medulloblastoma provided staging and radiation therapy are performed under optimal conditions.
INTRODUCTION
Medulloblastoma represents 15% to 20% of all brain tumors in children. Treatment comprising surgery followed by whole-CNS radiation therapy has significantly increased the survival rate, but causes major sequelae.
Standard-risk medulloblastoma is defined by strict criteria: total or subtotal tumor resection, no visible metastases on craniospinal magnetic resonance imaging (MRI), and no meningeal dissemination on postoperative lumbar puncture CSF cytology.1,2 Up until the 1990s, the reference postoperative treatment in this risk category was surgery followed by radiation therapy at 55 Gy to the posterior fossa and 35 Gy to the brain and spinal canal. Under these conditions, the 5-year recurrence-free survival rate was approximately 60%, but serious late effects were described. The most serious is neurocognitive impairment, causing difficulties of concentration, memorization, and social and school integration.3,4 Other sequelae are associated with treatment: primarily endocrine sequelae, especially those due to growth hormone deficiency.5 These sequelae, which mainly are related to craniospinal radiation therapy and are more frequent and severe in young children, have led to attempts to reduce the dose of prophylactic craniospinal radiation therapy.
This dose reduction appears to be particularly appropriate, given that a correlation clearly has been demonstrated between reduction of radiation doses and reduction of neurologic sequelae in children with leukemia.6,7 Since the 1990s, imaging conditions allowed a better identification of standard-risk patients, and several authors reported their experience of reduced-dose craniospinal radiation therapy in this patient category. However, a number of large-scale randomized trials have failed to formally demonstrate equivalent survival after reduced-dose craniospinal radiation therapy.8-10 Nevertheless, a correlation has been demonstrated between dose reduction and reduction of sequelae, particularly neurocognitive sequelae.11-14
In this study, the French Society of Pediatric Oncology (SFOP) Brain Tumor group wanted to evaluate the efficacy of a treatment comprising postoperative chemotherapy, which was effective in phase II trials, followed by reduced-dose craniospinal radiation therapy to try to reduce radiotherapy-related sequelae, without increasing the recurrence rate. The secondary objectives were to institute a review of inclusion criteria and radiation therapy planning and to conduct a detailed prospective study of sequelae.
PATIENTS AND METHODS
The diagnosis of medulloblastoma was based on histologic examination of the operative specimen from the initial operation on the posterior fossa tumor. Patients eligible for inclusion were between the ages of 3 and 18 years with a normal hematologic assessment (platelets > 100,000/mm3, WBC > 2,500/mm3, neutrophils > 1,000/mm3), normal renal function (creatinine clearance > 50 mL/min/1.73 m2), liver function tests showing parameters 0 or 1 according to the WHO toxicity scale, and a postoperative course allowing the initiation of chemotherapy. In view of the encouraging results obtained in young children treated with exclusive postoperative chemotherapy for standard-risk medulloblastoma, the lower age limit for inclusion was increased from 3 years old to 5 years old during the study. The preoperative assessment comprised clinical examination and cerebral imaging, preferably based on MRI. Spinal axis MRI was performed preoperatively when possible.
Early postoperative imaging after tumor resection comprised nonenhanced and contrast-enhanced brain MRI or computed tomography (CT), and spinal canal MRI when not performed preoperatively. In combination with the surgical findings, these images demonstrated the subtotal or total nature of the resection (subtotal resection was defined by the absence of identifiable residual tumor, whereas the surgeon described residual tumor, usually on the floor of the fourth ventricle) and the absence of metastasis. Postoperative lumbar puncture CSF cytology demonstrated the absence of tumor cells. All of these staging procedures were mandatory before study entry, and informed consent was required from parents or guardians. According to French law, the study was approved by an independent institutional review board.
The postoperative treatment protocol comprised chemotherapy followed by radiation therapy. Chemotherapy could start 1 week after surgery, when possible, and comprised the successive administration of two courses of eight drugs (vincristine, carmustine, methylprednisolone, procarbazine, cisplatine, cyclophosphamide, cytarabine, hydroxyurea) administered in 1 day (8/1),15-17 followed by two courses of etoposide (500 mg/m2) and carboplatin (800 mg/m2).18 This chemotherapy was not allowed to delay radiation therapy, which had to begin, at the latest, 105 days after surgery.
Scheduled hematologic surveillance was performed twice weekly, with a 2-week interval after each course of 8/1 and a 21-day interval after each course of etoposide and carboplatin. The chemotherapy schedule could be modified according to toxicity and the delay acquired between courses: only one 8/1 course when chemotherapy was started after postoperative day 30 and a reduced-dose second course of etoposide plus carboplatin according to the hematologic toxicity of the first course were allowed.
Cerebrospinal imaging as well as lumbar puncture CSF cytology were planned before radiation therapy. Any sign of progression on completion of chemotherapy was considered to represent protocol failure. In the absence of progression, radiation therapy consisted of 55 Gy to the posterior fossa and 25 Gy to the craniospinal axis. Radiation therapy was delivered in five sessions per week with fractions of 1.8 Gy per session.
Subsequent surveillance of disease-related and treatment-related adverse effects was performed regularly. A complete assessment of neurocognitive sequelae was initially planned, but was performed only in some centers. Multicenter follow-up was then performed by recording simplified data regarding sequelae; evaluating gait, writing, locomotor activity, and any change of laterality; and noting the presence or absence of epilepsy, cranial nerve lesions, or sensory impairment (vision, hearing). More specific neurocognitive impairment was investigated by evaluating school integration (normal, behind by > 2 years, special schooling), the intelligence quotient (IQ) for some children, and the need for special management (speech therapy, physiotherapy, psychological support). The SFOP Brain Tumor group initiated the use of a Health Utilities Index (HUI)19 adapted to children, translated into French, and validated.20 Clinical and biologic endocrine assessments were standard.5 Alopecia and its extent were also evaluated.
On the basis of an estimated 3-year survival rate of 70%, this open study had to include a minimum of 85 patients to achieve a 95% CI of ± 10%.
In parallel, a stopping rule was introduced to discontinue the study if the 1-year recurrence rate was greater than the acceptable limit. A triangular test methodology was adopted. The limits were calculated by considering a 1-year recurrence rate greater than 30% to be unacceptable, based on a reference 1-year recurrence rate of 10%, with error risks and of .05 and a frequency of analysis every 15 patients after a follow-up of 1 year. To take into account the real follow-up time, follow-up is calculated by the Kaplan-Meier method, inverting censored data and deaths. Survival and recurrence-free interval were estimated by the Kaplan-Meier method and were compared by the log-rank test. Rates are expressed together with 1.96 times the SE. Given that quality controls (neuropathology, radiology, and radiotherapy) were performed retrospectively, results are presented on an intention-to-treat basis.
Four neuropathologists reviewed the initial histologic documents to confirm the diagnosis of medulloblastoma. Four radiologists reviewed the preoperative and postoperative examinations and classified the data into four groups: correct inclusion, wrong inclusion (definite presence of residual tumor or metastases), doubtful images (doubt regarding the presence of residual tumor or axial metastases on craniospinal imaging), or missing data (axial MRI or postoperative brain MRI not reviewed or not provided at the time of the evaluation). Regular meetings of SFOP radiation oncologists were organized. The various dosimetric and imaging documents for each patient were studied every 6 months. Deviations were considered as major if safety margins less than 3 or 5 mm were observed, according the protocol guidelines.21
RESULTS
One hundred thirty-six patients (80 boys, 56 girls) were included between November 1991 and June 1998. The median age was 8 years (range, 3 to 18 years). Data were updated in June 2003, with a median follow-up of 6.5 years (range, 2 months to 10.5 years). For the 96 patients last known to be alive, the median time elapsed between last follow-up and June 30, 2003 is 17 months.
Brain MRI was performed in 131 children and CT only was performed in five children. The most frequent tumor site was the vermis in 75% of patients (98 of 131). Hydrocephalus was visible in 71% of patients (93 of 131).
A CSF internal shunt (usually ventriculoperitoneal shunt) was performed in 34 children, most of them preoperatively (30 of 34). The operative report was reviewed for 135 patients: tumor resection was considered to be total in 98 patients and subtotal in 37 patients. Invasion of the floor of the fourth ventricle was described by the neurosurgeon in 47% of patients (62 of 132). Histologic examination of the operative specimen confirmed the diagnosis of classical medulloblastoma in 116 patients and desmoplastic medulloblastoma in 20 patients. All patients underwent early postoperative imaging by CT and/or MRI, craniospinal MRI, and lumbar puncture CSF cytology that fit inclusion criteria.
The median interval between surgery and start of chemotherapy was 14 days (range 6 to 42 days). This interval was longer than 30 days in five patients. One hundred sixteen patients received platelet transfusion. Thirty-nine patients experienced 49 episodes of fever during grade 4 neutropenia. Dose reduction of the second course of etoposide and carboplatin was necessary in 33% of patients. There were no deaths as a result of toxicity. Eighty-two patients underwent complete assessment including craniospinal MRI and lumbar puncture CSF cytology before radiotherapy; 52 had incomplete assessment (49 of them had brain imaging) and two had no assessment. Among the 82 completely assessed patients, persistent complete remission was demonstrated in 96.3% of patients (n = 79), progressive disease was observed in two patients (one because of meningeal invasion detected on CSF cytology and one because of local relapse and meningeal invasion on CSF cytology), and residual tumor was detected in one patient (related to a residual tumor not initially identified on the postoperative assessment). Persistent complete remission was also recorded in 96% of the 52 incompletely assessed patients.
The median interval between surgery and the start of radiation therapy was 97 days. (range, 72 to 168 days). This interval was longer than 105 days for 37 patients because of treatment-related complications.
The median doses were 54 Gy to the posterior fossa and 25 Gy to the brain and the spinal axis. One patient wrongly received a craniospinal dose of 20 Gy. A dose of less than 28 Gy was delivered in 116 patients. Doses to the brain and to the spinal axis greater than 30 Gy were delivered to five patients because of age, modification of risk category, or progression before radiation therapy. Radiation therapy was associated with only minor hematologic toxicity. Whole-CNS MRI evaluation 2 months after the end of radiotherapy was performed in 126 patients: 117 were in complete remission (92.8%), one was in partial response on an initially unrecognized residual disease, one was in stable disease, and seven were in progressive disease.
Forty-seven recurrences have been recorded to date. Sites of relapses were posterior fossa only (n = 4), posterior fossa and distant CNS disseminated disease (n = 23, including one occurrence associated with extraneural disease), or CNS disseminated disease only (n = 20). In summary, 57% of the relapses included posterior fossa, 51% of the relapses included the spine, and 53% of the relapses included the brain. The criteria requiring discontinuation of the study on the basis of the 1-year recurrence rate were not met. The 5-year recurrence-free survival rate was 64.8% ± 8.1%. The overall 3-year survival rate was 81.3% ± 6.6%, compatible with the objectives of the trial. The 5-year overall survival was 73.8% ± 7.6%. These curves were based on the 136 patients included (Figs 1 and 2).
Eighty-seven sequelae forms were analyzed. These evaluations were performed between 1 and 11 years after the start of treatment (median, 6 years). The sequelae are listed in Table 1. Fifty-two patients completed the HUI self-administered questionnaire (ie, 54.2% of the patients alive at last follow-up). Their median follow-up, calculated from the date of surgery, was 8 years (range, 4.5 to 11 years). Their mean age was 17 years (range, 10 to 26). The levels for each attribute are described in Table 2. Overall, the levels on all eight attributes could be calculated for 45 patients, and 86.7% presented at least one altered attribute of the HUI-3 scale. Only one attribute was affected in 20% of patients, two attributes were affected in 35.6% of patients, and at least three attributes were affected in 31.1% of patients. As shown in Table 2, this impairment was usually scored as level 2 on a scale from 1 (normal) to 6 (severely altered), apart from cognition and speech.
Reviews
Histology was reviewed for 132 patients. Discordance was observed for one patient, in which the pathologists concluded that the patient had an atypical teratoid rhabdoid tumor. There had been no attempt to identify large-cell/anaplastic medulloblastoma.22 The 5-year recurrence-free survival rate in the 131 medulloblastomas was 65.7% ± 7.2% and the 5-year survival was 75% ± 7.7%.
Radiologic inclusion criteria were reviewed in 128 patients (94.1%). Ninety patients were correctly included, and four patients were wrongly included because they presented definite local residual tumor. Eighteen patients presented doubtful imaging, and the remaining 16 patients presented missing data, mainly craniospinal MRI images that were unavailable for review. As shown in Figures 3 and 4, no significant difference in prognosis was demonstrated according to radiologic review for the 112 assessable patients.
Radiotherapy planning documents were reviewed in 120 patients (88.2%). The radiation oncologists concluded the presence of a major deviation in 26 patients and at least two major deviations in 14 patients. Recurrence-free and overall survival according to radiotherapy deviation are presented in Figures 5 and 6, respectively.
In the 71 patients in whom the three reviews confirming the histologic diagnosis of medulloblastoma, the standard risk criterion, and the correct indication for radiation therapy were available, the 5-year recurrence-free survival rate was 71.8% ± 10.5%. The 3-year survival rate was 88.6% ± 7.5%, and the 5-year survival rate was 79.7% ± 9.5%.
DISCUSSION
In the 136 patients included in our study, the 5-year recurrence-free survival rate was 64.8% ± 8.1%, whereas the overall survival rate was 73.8% ± 7.6%. These results appear to be in agreement with the various published studies on populations treated by craniospinal radiation therapy at doses of 35 Gy, with 5-year recurrence-free survival rates range from 55% to 67%.8,23-25
The quality of staging, including good quality craniospinal MRI (to visualize all of the CNS and spinal canal), remains important because reduced-dose radiation therapy cannot be considered in children with high-risk medulloblastoma. In view of the uncertainty regarding the prognostic value of residual tumor and CSF, only children who have undergone total or subtotal resection with negative CSF cytology can be legitimately included in dose-reduction radiation therapy studies. However, several authors have defined residual tumor in different ways, making comparisons, as well as the interpretation of prognostic value, particularly complex. According to various authors, postoperative residual tumor is defined as the presence of a residual mass on early postoperative imaging with a largest surface area greater than 1.5 cm2,26 with a volume greater than 1.5 cm3,10 representing more than 50% of the initial tumor,12 or as the presence of any measurable residual tumor.1 The presence of residual tumor did not appear to influence prognosis in the International Society of Pediatric Oncology II study,8 whereas the study published by Zeltzer26 demonstrated the negative prognostic impact of residual tumor. The prognostic value of this parameter is all the more difficult to establish in that postoperative imaging is not always easy to interpret. Even with nonenhanced and contrast-enhanced early postoperative imaging, the distinction between residual tumor and postoperative scarring can remain difficult.
This difficulty of interpretation of imaging is illustrated by the retrospective identification by the investigators of a residual tumor in two children included in our study. The panel of radiologists also identified four wrongly included cases related to a missed residual tumor. The difficulties of staging are also illustrated by the results of radiologic review, which showed that 16 of the 128 imaging files were incomplete. In practice, when there is a doubt about the presence of metastasis on craniospinal MRI or when imaging is not complete, it is preferable to perform a second examination to determine the risk criteria clearly. Although the difference was not statistically significant, the 5-year recurrence-free survival rate was 69.8% for correctly included patients versus 52.5% for wrongly included patients. The absence of significant difference could have been due to the small number of patients. To improve staging, quality control could be performed prospectively, according to appropriate technique,27 reducing the risk of missing residual tumor or metastasis.
The prognostic value of CSF cytology is also a subject of debate. In our study, all postoperative lumbar puncture CSF cytology examinations were negative as defined by the inclusion criteria. However, at the time of assessment performed before radiotherapy was started, CSF progression was demonstrated in three children, in the absence of any other site of progression. Although CSF cytology was normal postoperatively, demonstration of tumor cells 90 days later raises a doubt about the true negativity of the initial CSF, given that the presence of tumor cells in the CSF can vary over time. A false-negative postoperative CSF examination cannot be formally excluded in these three occurrences of CSF progression. However, both negative MRI and CSF cytologic studies are necessary to confirm standard-risk medulloblastoma.28
The quality of craniospinal radiation therapy is also absolutely essential in dose-reduction regimens. Many criteria are involved, such as positioning of the patient, dose fractionation, definition of target volumes, the junction of various irradiation fields, and the choice of energy beams. In our study, the 5-year recurrence-free survival rate was 69.5% in the group of patients treated with zero or one major deviation, versus 50% when at least two major deviations were identified retrospectively. This difference was not statistically significant, possibly because of the small number of patients. Conversely, the difference in terms of 5-year overall survival rates was statistically significant, with a survival rate of 79.3% in the presence of zero or one major deviation and only 54.4% in the presence of at least two major deviations. Our results are in agreement with those published by Carrie,21 who studied the impact of major deviations on medulloblastomas and established a correlation between the presence of major deviation and unfavorable prognosis in patients treated for standard-risk (including some of the patients of our study) and also high-risk disease.
It is interesting to note that the quality of radiotherapy also has a specific prognostic value in the overall population of children with standard-risk medulloblastoma, in whom the dose of craniospinal radiation therapy was decreased to 25 Gy. In view of the technical complexity of radiation therapy and the demonstration of these major deviations, radiation therapy for medulloblastoma should probably be performed only in specialized referral centers. A prospective review of the technical conditions of planning and delivery of radiation therapy could be proposed before radiation to reduce the number of possible major deviations.
The pattern of relapse in our study did not significantly differ from that reported in recent series with higher doses of craniospinal radiotherapy.25,29 The results of single-center and multicenter nonrandomized studies of dose reduction in standard-risk medulloblastoma have shown recurrence-free survival rates ranging from 60% to 80%. However, these studies often included a limited number of patients.17,30-33
Two large-scale, multicenter, prospective, randomized studies evaluated the justification for reduced-dose craniospinal radiation therapy. In the Pediatric Oncology Group-Children’s Cancer Study Group study, the authors initially reported an increased rate of early spinal recurrences,9 but a second publication on the same population with a median follow-up of 8 years did not demonstrate any significant difference for recurrence-free survival between the two arms.10 In the International Society of Pediatric Oncology II study, the authors8 emphasized a negative interaction between chemotherapy and reduced-dose craniospinal radiation therapy: the lowest recurrence-free survival rate was observed in patients in whom reduced-dose craniospinal radiation therapy was administered after postoperative chemotherapy. Conversely, the group of patients treated by craniospinal radiation therapy at 25 Gy, but delivered just after the operation, presented a comparable recurrence-free survival rate to the group receiving radiation therapy at 35 Gy. The 5-year recurrence-free survival rate of patients included in our study with all optimal quality controls of histology, radiology, and radiotherapy also compares favorably to these series (71.8% ± 10.5%).
One multicenter, prospective, nonrandomized study, including 79 children with standard-risk medulloblastoma, has reported better results12: craniospinal radiation therapy was performed at reduced doses (23.4 Gy) and was associated with weekly injections of vincristine, followed by maintenance chemotherapy comprising lomustine, vincristine, and cisplatin. Sixty-five of the 79 patients included were considered to be eligible: 14 children were excluded, most of them because of problems of interpretation of staging. The 5-year recurrence-free survival rate of these 65 patients was 79% ± 7%. Subgroup analyses of our study did not demonstrate any significant difference for recurrence-free survival rate according to the quality of staging, but clearly demonstrated a negative impact of radiotherapy performed under suboptimal conditions. These results were confirmed in a recent preliminary report of a Children's Oncology Group multicentric study34 showing a three-year progression-free survival of approximately 85% in 322 eligible patients of 421 enrolled patients, who were treated with a 23.4 Gy craniospinal dose, local radiotherapy (54.8 Gy), and adjuvant chemotherapy. However, the authors emphasize the high incidence of ototoxicity in this study (25% of patients).
The benefit of chemotherapy in standard-risk medulloblastoma has been demonstrated only in one prospective randomized study, but all patients received a craniospinal radiation therapy dose of 35 Gy.25 However, all of the current results in standard-risk medulloblastoma, including those reported here, support a reduction of the craniospinal radiation dose from 35 to 25 Gy and the use of chemotherapy in combination with postoperative radiation therapy to decrease the risk of craniospinal and noncraniospinal metastasis. Theoretical advantages of preradiation chemotherapy compared with postradiation chemotherapy are a better brain penetration of the drugs just after surgery and a better hematologic tolerance before craniospinal radiotherapy. However, no study, including ours, has proven any benefit for patients treated with reduced-dose radiation therapy when chemotherapy is given before rather than after radiotherapy.
Prospective evaluation of cognitive sequelae proved to be difficult in our multicenter study. Several IQ measurements were available, but this sample is not representative, especially in view of the probable bias related to the various centers: some performed this evaluation on all children, whereas others only evaluated patients already presenting major neurocognitive impairment. Furthermore, IQ is probably most useful in the context of more complete neurocognitive assessment also investigating attention, temporospatial orientation, reasoning, writing, and language, which, in addition to precise assessment of other sequelae, allows more adapted management of these children to improve their integration. The use of HUI is advantageous in multicenter prospective studies because it is simple and reproducible: the results can be easily collected in patient cohorts.19,20 Furthermore, it can be useful for individual screening to propose more detailed investigations, at least in patients in whom cognitive impairment is detected.35 According to various studies, 30% to 65% of children are behind at school,14,35,36 which is another classical parameter for evaluation of impairment of higher functions. This disparity is probably further accentuated by differences in educational systems in different countries; the fact that a child is not behind at school is not always correlated with absence of cognitive sequelae. The need for special schooling appears to be a more reliable criterion. In our study, approximately 25% of children required special schooling, which is not significantly different from the rates reported in recent series, even after conventional-dose craniospinal radiation therapy.3,36
Of the 86 patients evaluated, 49% presented at least one disease-related or treatment-related neurologic impairment, especially an impairment that affected gait or writing. Finally, identification and management of neurosensory sequelae (visual or auditory) are also essential to ensure optimal integration. All of these sequelae must be evaluated prospectively not only to evaluate the effects of modifications of treatment regimens in homogeneous series of patients, but also to ensure optimal integration of each child by proposing the most adapted support.
Reduction of the craniospinal radiation dose, which is associated with a lower but still persistent risk of sequelae, did not compromise the survival of patients included in this study. Reduced-dose craniospinal radiation therapy can be proposed in standard-risk medulloblastoma provided staging and radiation therapy are performed under optimal conditions. It is hoped that additional biologic criteria37,38 will help to identify those patients that may benefit from additional decreased intensity of treatment and those who need to be more intensively treated.
Appendix
The following were coinvestigators of this study: H. Rubie, Y. Perel, A. Gofstein, F. Mechinaud, P. Tron, F. Dusol, A. Thyss, F. Millot, L. de Lumley, S. Bracard, D. Couanet, P. Thiesse, D. Figarella, A. Jouvet, A. Lellouch-Tubiana, C. Alapetite, M. Benhassel, V. Bernier, S. Chapet, J.P. Cuillere, F. Gomez, J.-L. Habrand, S. Hofstteter, H. Kolodie, J.-L. Lagrange, J.-P. Maire, X. Murraciole, P. Quetin, D. Bataille, G. Besson, S. Boetto, H. Boissonnet, P. Coubes, M. Dautheribes, P. Delhemmes, Y. Guegan, P. Kehrli, G. Lena, J. Maheu, J.-C. Marchal, P. Mercier, C. Mottolese, J-G. Passagia, A. Pierre-Kahn, F. Proust, D. Renier, C. Sainte-Rose, and M. Zerah.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank all of the investigators of this study: members of the review panels of neuropathology, radiology, and radiation oncology; neurosurgeons; and pediatric oncologists of the SFOP centers.
NOTES
Supported by a grant from the Association pour la Recherche sur le Cancer and sponsored by the Institut Curie, Paris, France.
Presented in part at the 10th International Symposium on Paediatric Neuro-Oncology, June 9-12, 2002, London, United Kingdom.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Gnekow AK: Recommendations of the Brain Tumor Subcommittee for the reporting of trials: SIOP Brain Tumor Subcommittee—International Society of Pediatric Oncology. Med Pediatr Oncol 24:104-108, 1995
Kühl J, Doz F, Taylor R: Embryonic tumours, in Walker DPG, Punt J (eds): Central Nervous System Tumours in Childhood. London, United Kingdom, Arnold, 2003, pp 314-330
Hoppe-Hirsch E, Brunet L, Laroussinie F, et al: Intellectual outcome in children with malignant tumors of the posterior fossa: Influence of the field of irradiation and quality of surgery. Childs Nerv Syst 11:340-346, 1995
Duffner PK, Cohen ME, Thomas P: Late effects of treatment on the intelligence of children with posterior fossa tumors. Cancer 51:233-237, 1983
Oberfield SE, Allen JC, Pollack J, et al: Long-term endocrine sequelae after treatment of medulloblastoma: Prospective study of growth and thyroid function. J Pediatr 108:219-223, 1986
Stubberfield TG, Byrne GC, Jones TW: Growth and growth hormone secretion after treatment for acute lymphoblastic leukemia in childhood: 18-Gy versus 24-Gy cranial irradiation. J Pediatr Hematol Oncol 17:167-171, 1995
Waber DP, Shapiro BL, Carpentieri SC, et al: Excellent therapeutic efficacy and minimal late neurotoxicity in children treated with 18 grays of cranial radiation therapy for high-risk acute lymphoblastic leukemia: A 7-year follow-up study of the Dana-Farber Cancer Institute Consortium Protocol 87-01. Cancer 92:15-22, 2001
Bailey CC, Gnekow A, Wellek S, et al: Prospective randomised trial of chemotherapy given before radiotherapy in childhood medulloblastoma: International Society of Paediatric Oncology (SIOP) and the (German) Society of Paediatric Oncology (GPO)—SIOP II. Med Pediatr Oncol 25:166-178, 1995
Deutsch M, Thomas PR, Krischer J, et al: Results of a prospective randomized trial comparing standard dose neuraxis irradiation (3,600 cGy/20) with reduced neuraxis irradiation (2,340 cGy/13) in patients with low-stage medulloblastoma: A combined Children's Cancer Group-Pediatric Oncology Group Study. Pediatr Neurosurg 24:167-177, 1996
Thomas PR, Deutsch M, Kepner JL, et al: Low-stage medulloblastoma: Final analysis of trial comparing standard-dose with reduced-dose neuraxis irradiation. J Clin Oncol 18:3004-3011, 2000
Mulhern RK, Kepner JL, Thomas PR, et al: Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced-dose craniospinal irradiation: A Pediatric Oncology Group study. J Clin Oncol 16:1723-1728, 1998
Packer RJ, Goldwein J, Nicholson HS, et al: Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: A Children's Cancer Group Study. J Clin Oncol 17:2127-2136, 1999
Ris MD, Packer R, Goldwein J, et al: Intellectual outcome after reduced-dose radiation therapy plus adjuvant chemotherapy for medulloblastoma: A Children's Cancer Group study. J Clin Oncol 19:3470-3476, 2001
Grill J, Renaux VK, Bulteau C, et al: Long-term intellectual outcome in children with posterior fossa tumors according to radiation doses and volumes. Int J Radiat Oncol Biol Phys 45:137-145, 1999
Pendergrass TW, Milstein JM, Geyer JR, et al: Eight drugs in one day chemotherapy for brain tumors: Experience in 107 children and rationale for preradiation chemotherapy. J Clin Oncol 5:1221-1231, 1987
Chastagner P, Olive D, Philip T, et al: Efficacy of the "8 drugs in a day" protocol in brain tumors in children. Arch Fr Pediatr 45:249-254, 1988
Gentet JC, Bouffet E, Doz F, et al: Preirradiation chemotherapy including "eight drugs in 1 day" regimen and high-dose methotrexate in childhood medulloblastoma: Results of the M7 French Cooperative Study. J Neurosurg 82:608-614, 1995
Gentet JC, Doz F, Bouffet E, et al: Carboplatin and VP 16 in medulloblastoma: A phase II study of the French Society of Pediatric Oncology (SFOP). Med Pediatr Oncol 23:422-427, 1994
Feeny D, Furlong W, Boyle M, et al: Multi-attribute health status classification systems: Health Utilities Index. Pharmacoeconomics 7:490-502, 1995
Le Gales C, Costet N, Gentet JC, et al: Cross-cultural adaptation of a health status classification system in children with cancer: First results of the French adaptation of the Health Utilities Index Marks 2 and 3. Int J Cancer Suppl 12:112-118, 1999
Carrie C, Hoffstetter S, Gomez F, 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 45:435-439, 1999
Eberhart CG, Burger PC: Anaplasia and grading in medulloblastomas. Brain Pathol 13:376-385, 2003
Tait DM, Thornton-Jones H, Bloom HJ, et al: Adjuvant chemotherapy for medulloblastoma: The first multi-centre control trial of the International Society of Paediatric Oncology (SIOP I). Eur J Cancer 26:464-469, 1990
Evans AE, Jenkin RD, Sposto R, et al: The treatment of medulloblastoma: Results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine, and prednisone. J Neurosurg 72:572-582, 1990
Taylor RE, Bailey CC, Robinson K, et al: Results of a randomized study of preradiation chemotherapy versus radiotherapy alone for nonmetastatic medulloblastoma: The International Society of Paediatric Oncology/United Kingdom Children's Cancer Study Group PNET-3 Study. J Clin Oncol 21:1581-1591, 2003
Zeltzer PM, Boyett JM, Finlay JL, 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 17:832-845, 1999
Thiesse P, Jaspan T, Couanet D, et al: A protocol for imaging pediatric brain tumors. J Radiol 82:11-16, 2001
Fouladi M, Gajjar A, Boyett JM, et al: Comparison of CSF cytology and spinal magnetic resonance imaging in the detection of leptomeningeal disease in pediatric medulloblastoma or primitive neuroectodermal tumor. J Clin Oncol 17:3234-3237, 1999
Kortmann 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 46:269-279, 2000
Brand WN, Schneider PA, Tokars RP: Long-term results of a pilot study of low dose cranial-spinal irradiation for cerebellar medulloblastoma. Int J Radiat Oncol Biol Phys 13:1641-1645, 1987
Halberg FE, Wara WM, Fippin LF, et al: Low-dose craniospinal radiation therapy for medulloblastoma. Int J Radiat Oncol Biol Phys 20:651-654, 1991
Levin VA, Rodriguez LA, Edwards MS, et al: Treatment of medulloblastoma with procarbazine, hydroxyurea, and reduced radiation doses to whole brain and spine. J Neurosurg 68:383-387, 1988
Hughes EN, Shillito J, Sallan SE, et al: Medulloblastoma at the Joint Center for Radiation Therapy between 1968 and 1984: The influence of radiation dose on the patterns of failure and survival. Cancer 61:1992-1998, 1988
Packer RJ, Gajjar R, Vezina G, et al: Preliminary results of a prospective randomized study of "reduced" craniospinal radiotherapy followed by one of two adjuvant chemotherapy regimens for children with medulloblastoma: Children's Oncology Group study. Ann Neurol 56:s89-s90, 2004 (suppl 8)
Mulhern RK: Correlation of the Health Utilities Index Mark 2 cognition scale and neuropsychological functioning among survivors of childhood medulloblastoma. Int J Cancer Suppl 12:91-94, 1999
Palmer SL, Goloubeva O, Reddick WE, et al: Patterns of intellectual development among survivors of pediatric medulloblastoma: A longitudinal analysis. J Clin Oncol 19:2302-2308, 2001
Gajjar A, Hernan R, Kocak M, et al: Clinical, histopathologic, and molecular markers of prognosis: Toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 22:984-993, 2004
Fernandez-Teijeiro A, Betensky RA, Sturla LM, et al: Combining gene expression profiles and clinical parameters for risk stratification in medulloblastomas. J Clin Oncol 22:994-998, 2004(V. Oyharcabal-Bourden, C.)
Institut Gustave Roussy, Villejuif
Centre Hospitalo–Universitaire (CHU) La Timone, Marseille
Centre Leon Berard, Lyon
CHU Rennes, CHU Nancy
CHU Grenoble
CHU Strasbourg
Centre Oscar Lambret, Lille
CHU Lille
Unite de l'Institut National de la Sante et de la Recherche Medicale U537, Le Kremlin Bicêtre, France
Hpital Reine Fabiola, Brussels, Belgium
ABSTRACT
OBJECTIVE: The primary objective of this study was to decrease the late effects of prophylactic radiation without reducing survival in standard-risk childhood medulloblastoma.
PATIENTS AND METHODS: Inclusion criteria were as follows: children between the ages of 3 and 18 years with total or subtotal tumor resection, no metastasis, and negative postoperative lumbar puncture CSF cytology. Two courses of eight drugs in 1 day followed by two courses of etoposide plus carboplatin (500 and 800 mg/m2 per course, respectively) were administered after surgery. Radiation therapy had to begin 90 days after surgery. Delivered doses were 55 Gy to the posterior fossa and 25 Gy to the brain and spinal canal.
RESULTS: Between November 1991 and June 1998, 136 patients (median age, 8 years; median follow-up, 6.5 years) were included. The overall survival rate and 5-year recurrence-free survival rate were 73.8% ± 7.6% and 64.8% ± 8.1%, respectively. Radiologic review showed that 4% of patients were wrongly included. Review of radiotherapy technical files demonstrated a correlation between the presence of a major protocol deviation and treatment failure. The 5-year recurrence-free survival rate of patients included in this study with all optimal quality controls of histology, radiology, and radiotherapy was 71.8% ± 10.5%. In terms of sequelae, 31% of patients required growth hormone replacement therapy and 25% required special schooling.
CONCLUSION: Reduced-dose craniospinal radiation therapy can be proposed in standard-risk medulloblastoma provided staging and radiation therapy are performed under optimal conditions.
INTRODUCTION
Medulloblastoma represents 15% to 20% of all brain tumors in children. Treatment comprising surgery followed by whole-CNS radiation therapy has significantly increased the survival rate, but causes major sequelae.
Standard-risk medulloblastoma is defined by strict criteria: total or subtotal tumor resection, no visible metastases on craniospinal magnetic resonance imaging (MRI), and no meningeal dissemination on postoperative lumbar puncture CSF cytology.1,2 Up until the 1990s, the reference postoperative treatment in this risk category was surgery followed by radiation therapy at 55 Gy to the posterior fossa and 35 Gy to the brain and spinal canal. Under these conditions, the 5-year recurrence-free survival rate was approximately 60%, but serious late effects were described. The most serious is neurocognitive impairment, causing difficulties of concentration, memorization, and social and school integration.3,4 Other sequelae are associated with treatment: primarily endocrine sequelae, especially those due to growth hormone deficiency.5 These sequelae, which mainly are related to craniospinal radiation therapy and are more frequent and severe in young children, have led to attempts to reduce the dose of prophylactic craniospinal radiation therapy.
This dose reduction appears to be particularly appropriate, given that a correlation clearly has been demonstrated between reduction of radiation doses and reduction of neurologic sequelae in children with leukemia.6,7 Since the 1990s, imaging conditions allowed a better identification of standard-risk patients, and several authors reported their experience of reduced-dose craniospinal radiation therapy in this patient category. However, a number of large-scale randomized trials have failed to formally demonstrate equivalent survival after reduced-dose craniospinal radiation therapy.8-10 Nevertheless, a correlation has been demonstrated between dose reduction and reduction of sequelae, particularly neurocognitive sequelae.11-14
In this study, the French Society of Pediatric Oncology (SFOP) Brain Tumor group wanted to evaluate the efficacy of a treatment comprising postoperative chemotherapy, which was effective in phase II trials, followed by reduced-dose craniospinal radiation therapy to try to reduce radiotherapy-related sequelae, without increasing the recurrence rate. The secondary objectives were to institute a review of inclusion criteria and radiation therapy planning and to conduct a detailed prospective study of sequelae.
PATIENTS AND METHODS
The diagnosis of medulloblastoma was based on histologic examination of the operative specimen from the initial operation on the posterior fossa tumor. Patients eligible for inclusion were between the ages of 3 and 18 years with a normal hematologic assessment (platelets > 100,000/mm3, WBC > 2,500/mm3, neutrophils > 1,000/mm3), normal renal function (creatinine clearance > 50 mL/min/1.73 m2), liver function tests showing parameters 0 or 1 according to the WHO toxicity scale, and a postoperative course allowing the initiation of chemotherapy. In view of the encouraging results obtained in young children treated with exclusive postoperative chemotherapy for standard-risk medulloblastoma, the lower age limit for inclusion was increased from 3 years old to 5 years old during the study. The preoperative assessment comprised clinical examination and cerebral imaging, preferably based on MRI. Spinal axis MRI was performed preoperatively when possible.
Early postoperative imaging after tumor resection comprised nonenhanced and contrast-enhanced brain MRI or computed tomography (CT), and spinal canal MRI when not performed preoperatively. In combination with the surgical findings, these images demonstrated the subtotal or total nature of the resection (subtotal resection was defined by the absence of identifiable residual tumor, whereas the surgeon described residual tumor, usually on the floor of the fourth ventricle) and the absence of metastasis. Postoperative lumbar puncture CSF cytology demonstrated the absence of tumor cells. All of these staging procedures were mandatory before study entry, and informed consent was required from parents or guardians. According to French law, the study was approved by an independent institutional review board.
The postoperative treatment protocol comprised chemotherapy followed by radiation therapy. Chemotherapy could start 1 week after surgery, when possible, and comprised the successive administration of two courses of eight drugs (vincristine, carmustine, methylprednisolone, procarbazine, cisplatine, cyclophosphamide, cytarabine, hydroxyurea) administered in 1 day (8/1),15-17 followed by two courses of etoposide (500 mg/m2) and carboplatin (800 mg/m2).18 This chemotherapy was not allowed to delay radiation therapy, which had to begin, at the latest, 105 days after surgery.
Scheduled hematologic surveillance was performed twice weekly, with a 2-week interval after each course of 8/1 and a 21-day interval after each course of etoposide and carboplatin. The chemotherapy schedule could be modified according to toxicity and the delay acquired between courses: only one 8/1 course when chemotherapy was started after postoperative day 30 and a reduced-dose second course of etoposide plus carboplatin according to the hematologic toxicity of the first course were allowed.
Cerebrospinal imaging as well as lumbar puncture CSF cytology were planned before radiation therapy. Any sign of progression on completion of chemotherapy was considered to represent protocol failure. In the absence of progression, radiation therapy consisted of 55 Gy to the posterior fossa and 25 Gy to the craniospinal axis. Radiation therapy was delivered in five sessions per week with fractions of 1.8 Gy per session.
Subsequent surveillance of disease-related and treatment-related adverse effects was performed regularly. A complete assessment of neurocognitive sequelae was initially planned, but was performed only in some centers. Multicenter follow-up was then performed by recording simplified data regarding sequelae; evaluating gait, writing, locomotor activity, and any change of laterality; and noting the presence or absence of epilepsy, cranial nerve lesions, or sensory impairment (vision, hearing). More specific neurocognitive impairment was investigated by evaluating school integration (normal, behind by > 2 years, special schooling), the intelligence quotient (IQ) for some children, and the need for special management (speech therapy, physiotherapy, psychological support). The SFOP Brain Tumor group initiated the use of a Health Utilities Index (HUI)19 adapted to children, translated into French, and validated.20 Clinical and biologic endocrine assessments were standard.5 Alopecia and its extent were also evaluated.
On the basis of an estimated 3-year survival rate of 70%, this open study had to include a minimum of 85 patients to achieve a 95% CI of ± 10%.
In parallel, a stopping rule was introduced to discontinue the study if the 1-year recurrence rate was greater than the acceptable limit. A triangular test methodology was adopted. The limits were calculated by considering a 1-year recurrence rate greater than 30% to be unacceptable, based on a reference 1-year recurrence rate of 10%, with error risks and of .05 and a frequency of analysis every 15 patients after a follow-up of 1 year. To take into account the real follow-up time, follow-up is calculated by the Kaplan-Meier method, inverting censored data and deaths. Survival and recurrence-free interval were estimated by the Kaplan-Meier method and were compared by the log-rank test. Rates are expressed together with 1.96 times the SE. Given that quality controls (neuropathology, radiology, and radiotherapy) were performed retrospectively, results are presented on an intention-to-treat basis.
Four neuropathologists reviewed the initial histologic documents to confirm the diagnosis of medulloblastoma. Four radiologists reviewed the preoperative and postoperative examinations and classified the data into four groups: correct inclusion, wrong inclusion (definite presence of residual tumor or metastases), doubtful images (doubt regarding the presence of residual tumor or axial metastases on craniospinal imaging), or missing data (axial MRI or postoperative brain MRI not reviewed or not provided at the time of the evaluation). Regular meetings of SFOP radiation oncologists were organized. The various dosimetric and imaging documents for each patient were studied every 6 months. Deviations were considered as major if safety margins less than 3 or 5 mm were observed, according the protocol guidelines.21
RESULTS
One hundred thirty-six patients (80 boys, 56 girls) were included between November 1991 and June 1998. The median age was 8 years (range, 3 to 18 years). Data were updated in June 2003, with a median follow-up of 6.5 years (range, 2 months to 10.5 years). For the 96 patients last known to be alive, the median time elapsed between last follow-up and June 30, 2003 is 17 months.
Brain MRI was performed in 131 children and CT only was performed in five children. The most frequent tumor site was the vermis in 75% of patients (98 of 131). Hydrocephalus was visible in 71% of patients (93 of 131).
A CSF internal shunt (usually ventriculoperitoneal shunt) was performed in 34 children, most of them preoperatively (30 of 34). The operative report was reviewed for 135 patients: tumor resection was considered to be total in 98 patients and subtotal in 37 patients. Invasion of the floor of the fourth ventricle was described by the neurosurgeon in 47% of patients (62 of 132). Histologic examination of the operative specimen confirmed the diagnosis of classical medulloblastoma in 116 patients and desmoplastic medulloblastoma in 20 patients. All patients underwent early postoperative imaging by CT and/or MRI, craniospinal MRI, and lumbar puncture CSF cytology that fit inclusion criteria.
The median interval between surgery and start of chemotherapy was 14 days (range 6 to 42 days). This interval was longer than 30 days in five patients. One hundred sixteen patients received platelet transfusion. Thirty-nine patients experienced 49 episodes of fever during grade 4 neutropenia. Dose reduction of the second course of etoposide and carboplatin was necessary in 33% of patients. There were no deaths as a result of toxicity. Eighty-two patients underwent complete assessment including craniospinal MRI and lumbar puncture CSF cytology before radiotherapy; 52 had incomplete assessment (49 of them had brain imaging) and two had no assessment. Among the 82 completely assessed patients, persistent complete remission was demonstrated in 96.3% of patients (n = 79), progressive disease was observed in two patients (one because of meningeal invasion detected on CSF cytology and one because of local relapse and meningeal invasion on CSF cytology), and residual tumor was detected in one patient (related to a residual tumor not initially identified on the postoperative assessment). Persistent complete remission was also recorded in 96% of the 52 incompletely assessed patients.
The median interval between surgery and the start of radiation therapy was 97 days. (range, 72 to 168 days). This interval was longer than 105 days for 37 patients because of treatment-related complications.
The median doses were 54 Gy to the posterior fossa and 25 Gy to the brain and the spinal axis. One patient wrongly received a craniospinal dose of 20 Gy. A dose of less than 28 Gy was delivered in 116 patients. Doses to the brain and to the spinal axis greater than 30 Gy were delivered to five patients because of age, modification of risk category, or progression before radiation therapy. Radiation therapy was associated with only minor hematologic toxicity. Whole-CNS MRI evaluation 2 months after the end of radiotherapy was performed in 126 patients: 117 were in complete remission (92.8%), one was in partial response on an initially unrecognized residual disease, one was in stable disease, and seven were in progressive disease.
Forty-seven recurrences have been recorded to date. Sites of relapses were posterior fossa only (n = 4), posterior fossa and distant CNS disseminated disease (n = 23, including one occurrence associated with extraneural disease), or CNS disseminated disease only (n = 20). In summary, 57% of the relapses included posterior fossa, 51% of the relapses included the spine, and 53% of the relapses included the brain. The criteria requiring discontinuation of the study on the basis of the 1-year recurrence rate were not met. The 5-year recurrence-free survival rate was 64.8% ± 8.1%. The overall 3-year survival rate was 81.3% ± 6.6%, compatible with the objectives of the trial. The 5-year overall survival was 73.8% ± 7.6%. These curves were based on the 136 patients included (Figs 1 and 2).
Eighty-seven sequelae forms were analyzed. These evaluations were performed between 1 and 11 years after the start of treatment (median, 6 years). The sequelae are listed in Table 1. Fifty-two patients completed the HUI self-administered questionnaire (ie, 54.2% of the patients alive at last follow-up). Their median follow-up, calculated from the date of surgery, was 8 years (range, 4.5 to 11 years). Their mean age was 17 years (range, 10 to 26). The levels for each attribute are described in Table 2. Overall, the levels on all eight attributes could be calculated for 45 patients, and 86.7% presented at least one altered attribute of the HUI-3 scale. Only one attribute was affected in 20% of patients, two attributes were affected in 35.6% of patients, and at least three attributes were affected in 31.1% of patients. As shown in Table 2, this impairment was usually scored as level 2 on a scale from 1 (normal) to 6 (severely altered), apart from cognition and speech.
Reviews
Histology was reviewed for 132 patients. Discordance was observed for one patient, in which the pathologists concluded that the patient had an atypical teratoid rhabdoid tumor. There had been no attempt to identify large-cell/anaplastic medulloblastoma.22 The 5-year recurrence-free survival rate in the 131 medulloblastomas was 65.7% ± 7.2% and the 5-year survival was 75% ± 7.7%.
Radiologic inclusion criteria were reviewed in 128 patients (94.1%). Ninety patients were correctly included, and four patients were wrongly included because they presented definite local residual tumor. Eighteen patients presented doubtful imaging, and the remaining 16 patients presented missing data, mainly craniospinal MRI images that were unavailable for review. As shown in Figures 3 and 4, no significant difference in prognosis was demonstrated according to radiologic review for the 112 assessable patients.
Radiotherapy planning documents were reviewed in 120 patients (88.2%). The radiation oncologists concluded the presence of a major deviation in 26 patients and at least two major deviations in 14 patients. Recurrence-free and overall survival according to radiotherapy deviation are presented in Figures 5 and 6, respectively.
In the 71 patients in whom the three reviews confirming the histologic diagnosis of medulloblastoma, the standard risk criterion, and the correct indication for radiation therapy were available, the 5-year recurrence-free survival rate was 71.8% ± 10.5%. The 3-year survival rate was 88.6% ± 7.5%, and the 5-year survival rate was 79.7% ± 9.5%.
DISCUSSION
In the 136 patients included in our study, the 5-year recurrence-free survival rate was 64.8% ± 8.1%, whereas the overall survival rate was 73.8% ± 7.6%. These results appear to be in agreement with the various published studies on populations treated by craniospinal radiation therapy at doses of 35 Gy, with 5-year recurrence-free survival rates range from 55% to 67%.8,23-25
The quality of staging, including good quality craniospinal MRI (to visualize all of the CNS and spinal canal), remains important because reduced-dose radiation therapy cannot be considered in children with high-risk medulloblastoma. In view of the uncertainty regarding the prognostic value of residual tumor and CSF, only children who have undergone total or subtotal resection with negative CSF cytology can be legitimately included in dose-reduction radiation therapy studies. However, several authors have defined residual tumor in different ways, making comparisons, as well as the interpretation of prognostic value, particularly complex. According to various authors, postoperative residual tumor is defined as the presence of a residual mass on early postoperative imaging with a largest surface area greater than 1.5 cm2,26 with a volume greater than 1.5 cm3,10 representing more than 50% of the initial tumor,12 or as the presence of any measurable residual tumor.1 The presence of residual tumor did not appear to influence prognosis in the International Society of Pediatric Oncology II study,8 whereas the study published by Zeltzer26 demonstrated the negative prognostic impact of residual tumor. The prognostic value of this parameter is all the more difficult to establish in that postoperative imaging is not always easy to interpret. Even with nonenhanced and contrast-enhanced early postoperative imaging, the distinction between residual tumor and postoperative scarring can remain difficult.
This difficulty of interpretation of imaging is illustrated by the retrospective identification by the investigators of a residual tumor in two children included in our study. The panel of radiologists also identified four wrongly included cases related to a missed residual tumor. The difficulties of staging are also illustrated by the results of radiologic review, which showed that 16 of the 128 imaging files were incomplete. In practice, when there is a doubt about the presence of metastasis on craniospinal MRI or when imaging is not complete, it is preferable to perform a second examination to determine the risk criteria clearly. Although the difference was not statistically significant, the 5-year recurrence-free survival rate was 69.8% for correctly included patients versus 52.5% for wrongly included patients. The absence of significant difference could have been due to the small number of patients. To improve staging, quality control could be performed prospectively, according to appropriate technique,27 reducing the risk of missing residual tumor or metastasis.
The prognostic value of CSF cytology is also a subject of debate. In our study, all postoperative lumbar puncture CSF cytology examinations were negative as defined by the inclusion criteria. However, at the time of assessment performed before radiotherapy was started, CSF progression was demonstrated in three children, in the absence of any other site of progression. Although CSF cytology was normal postoperatively, demonstration of tumor cells 90 days later raises a doubt about the true negativity of the initial CSF, given that the presence of tumor cells in the CSF can vary over time. A false-negative postoperative CSF examination cannot be formally excluded in these three occurrences of CSF progression. However, both negative MRI and CSF cytologic studies are necessary to confirm standard-risk medulloblastoma.28
The quality of craniospinal radiation therapy is also absolutely essential in dose-reduction regimens. Many criteria are involved, such as positioning of the patient, dose fractionation, definition of target volumes, the junction of various irradiation fields, and the choice of energy beams. In our study, the 5-year recurrence-free survival rate was 69.5% in the group of patients treated with zero or one major deviation, versus 50% when at least two major deviations were identified retrospectively. This difference was not statistically significant, possibly because of the small number of patients. Conversely, the difference in terms of 5-year overall survival rates was statistically significant, with a survival rate of 79.3% in the presence of zero or one major deviation and only 54.4% in the presence of at least two major deviations. Our results are in agreement with those published by Carrie,21 who studied the impact of major deviations on medulloblastomas and established a correlation between the presence of major deviation and unfavorable prognosis in patients treated for standard-risk (including some of the patients of our study) and also high-risk disease.
It is interesting to note that the quality of radiotherapy also has a specific prognostic value in the overall population of children with standard-risk medulloblastoma, in whom the dose of craniospinal radiation therapy was decreased to 25 Gy. In view of the technical complexity of radiation therapy and the demonstration of these major deviations, radiation therapy for medulloblastoma should probably be performed only in specialized referral centers. A prospective review of the technical conditions of planning and delivery of radiation therapy could be proposed before radiation to reduce the number of possible major deviations.
The pattern of relapse in our study did not significantly differ from that reported in recent series with higher doses of craniospinal radiotherapy.25,29 The results of single-center and multicenter nonrandomized studies of dose reduction in standard-risk medulloblastoma have shown recurrence-free survival rates ranging from 60% to 80%. However, these studies often included a limited number of patients.17,30-33
Two large-scale, multicenter, prospective, randomized studies evaluated the justification for reduced-dose craniospinal radiation therapy. In the Pediatric Oncology Group-Children’s Cancer Study Group study, the authors initially reported an increased rate of early spinal recurrences,9 but a second publication on the same population with a median follow-up of 8 years did not demonstrate any significant difference for recurrence-free survival between the two arms.10 In the International Society of Pediatric Oncology II study, the authors8 emphasized a negative interaction between chemotherapy and reduced-dose craniospinal radiation therapy: the lowest recurrence-free survival rate was observed in patients in whom reduced-dose craniospinal radiation therapy was administered after postoperative chemotherapy. Conversely, the group of patients treated by craniospinal radiation therapy at 25 Gy, but delivered just after the operation, presented a comparable recurrence-free survival rate to the group receiving radiation therapy at 35 Gy. The 5-year recurrence-free survival rate of patients included in our study with all optimal quality controls of histology, radiology, and radiotherapy also compares favorably to these series (71.8% ± 10.5%).
One multicenter, prospective, nonrandomized study, including 79 children with standard-risk medulloblastoma, has reported better results12: craniospinal radiation therapy was performed at reduced doses (23.4 Gy) and was associated with weekly injections of vincristine, followed by maintenance chemotherapy comprising lomustine, vincristine, and cisplatin. Sixty-five of the 79 patients included were considered to be eligible: 14 children were excluded, most of them because of problems of interpretation of staging. The 5-year recurrence-free survival rate of these 65 patients was 79% ± 7%. Subgroup analyses of our study did not demonstrate any significant difference for recurrence-free survival rate according to the quality of staging, but clearly demonstrated a negative impact of radiotherapy performed under suboptimal conditions. These results were confirmed in a recent preliminary report of a Children's Oncology Group multicentric study34 showing a three-year progression-free survival of approximately 85% in 322 eligible patients of 421 enrolled patients, who were treated with a 23.4 Gy craniospinal dose, local radiotherapy (54.8 Gy), and adjuvant chemotherapy. However, the authors emphasize the high incidence of ototoxicity in this study (25% of patients).
The benefit of chemotherapy in standard-risk medulloblastoma has been demonstrated only in one prospective randomized study, but all patients received a craniospinal radiation therapy dose of 35 Gy.25 However, all of the current results in standard-risk medulloblastoma, including those reported here, support a reduction of the craniospinal radiation dose from 35 to 25 Gy and the use of chemotherapy in combination with postoperative radiation therapy to decrease the risk of craniospinal and noncraniospinal metastasis. Theoretical advantages of preradiation chemotherapy compared with postradiation chemotherapy are a better brain penetration of the drugs just after surgery and a better hematologic tolerance before craniospinal radiotherapy. However, no study, including ours, has proven any benefit for patients treated with reduced-dose radiation therapy when chemotherapy is given before rather than after radiotherapy.
Prospective evaluation of cognitive sequelae proved to be difficult in our multicenter study. Several IQ measurements were available, but this sample is not representative, especially in view of the probable bias related to the various centers: some performed this evaluation on all children, whereas others only evaluated patients already presenting major neurocognitive impairment. Furthermore, IQ is probably most useful in the context of more complete neurocognitive assessment also investigating attention, temporospatial orientation, reasoning, writing, and language, which, in addition to precise assessment of other sequelae, allows more adapted management of these children to improve their integration. The use of HUI is advantageous in multicenter prospective studies because it is simple and reproducible: the results can be easily collected in patient cohorts.19,20 Furthermore, it can be useful for individual screening to propose more detailed investigations, at least in patients in whom cognitive impairment is detected.35 According to various studies, 30% to 65% of children are behind at school,14,35,36 which is another classical parameter for evaluation of impairment of higher functions. This disparity is probably further accentuated by differences in educational systems in different countries; the fact that a child is not behind at school is not always correlated with absence of cognitive sequelae. The need for special schooling appears to be a more reliable criterion. In our study, approximately 25% of children required special schooling, which is not significantly different from the rates reported in recent series, even after conventional-dose craniospinal radiation therapy.3,36
Of the 86 patients evaluated, 49% presented at least one disease-related or treatment-related neurologic impairment, especially an impairment that affected gait or writing. Finally, identification and management of neurosensory sequelae (visual or auditory) are also essential to ensure optimal integration. All of these sequelae must be evaluated prospectively not only to evaluate the effects of modifications of treatment regimens in homogeneous series of patients, but also to ensure optimal integration of each child by proposing the most adapted support.
Reduction of the craniospinal radiation dose, which is associated with a lower but still persistent risk of sequelae, did not compromise the survival of patients included in this study. Reduced-dose craniospinal radiation therapy can be proposed in standard-risk medulloblastoma provided staging and radiation therapy are performed under optimal conditions. It is hoped that additional biologic criteria37,38 will help to identify those patients that may benefit from additional decreased intensity of treatment and those who need to be more intensively treated.
Appendix
The following were coinvestigators of this study: H. Rubie, Y. Perel, A. Gofstein, F. Mechinaud, P. Tron, F. Dusol, A. Thyss, F. Millot, L. de Lumley, S. Bracard, D. Couanet, P. Thiesse, D. Figarella, A. Jouvet, A. Lellouch-Tubiana, C. Alapetite, M. Benhassel, V. Bernier, S. Chapet, J.P. Cuillere, F. Gomez, J.-L. Habrand, S. Hofstteter, H. Kolodie, J.-L. Lagrange, J.-P. Maire, X. Murraciole, P. Quetin, D. Bataille, G. Besson, S. Boetto, H. Boissonnet, P. Coubes, M. Dautheribes, P. Delhemmes, Y. Guegan, P. Kehrli, G. Lena, J. Maheu, J.-C. Marchal, P. Mercier, C. Mottolese, J-G. Passagia, A. Pierre-Kahn, F. Proust, D. Renier, C. Sainte-Rose, and M. Zerah.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank all of the investigators of this study: members of the review panels of neuropathology, radiology, and radiation oncology; neurosurgeons; and pediatric oncologists of the SFOP centers.
NOTES
Supported by a grant from the Association pour la Recherche sur le Cancer and sponsored by the Institut Curie, Paris, France.
Presented in part at the 10th International Symposium on Paediatric Neuro-Oncology, June 9-12, 2002, London, United Kingdom.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Gnekow AK: Recommendations of the Brain Tumor Subcommittee for the reporting of trials: SIOP Brain Tumor Subcommittee—International Society of Pediatric Oncology. Med Pediatr Oncol 24:104-108, 1995
Kühl J, Doz F, Taylor R: Embryonic tumours, in Walker DPG, Punt J (eds): Central Nervous System Tumours in Childhood. London, United Kingdom, Arnold, 2003, pp 314-330
Hoppe-Hirsch E, Brunet L, Laroussinie F, et al: Intellectual outcome in children with malignant tumors of the posterior fossa: Influence of the field of irradiation and quality of surgery. Childs Nerv Syst 11:340-346, 1995
Duffner PK, Cohen ME, Thomas P: Late effects of treatment on the intelligence of children with posterior fossa tumors. Cancer 51:233-237, 1983
Oberfield SE, Allen JC, Pollack J, et al: Long-term endocrine sequelae after treatment of medulloblastoma: Prospective study of growth and thyroid function. J Pediatr 108:219-223, 1986
Stubberfield TG, Byrne GC, Jones TW: Growth and growth hormone secretion after treatment for acute lymphoblastic leukemia in childhood: 18-Gy versus 24-Gy cranial irradiation. J Pediatr Hematol Oncol 17:167-171, 1995
Waber DP, Shapiro BL, Carpentieri SC, et al: Excellent therapeutic efficacy and minimal late neurotoxicity in children treated with 18 grays of cranial radiation therapy for high-risk acute lymphoblastic leukemia: A 7-year follow-up study of the Dana-Farber Cancer Institute Consortium Protocol 87-01. Cancer 92:15-22, 2001
Bailey CC, Gnekow A, Wellek S, et al: Prospective randomised trial of chemotherapy given before radiotherapy in childhood medulloblastoma: International Society of Paediatric Oncology (SIOP) and the (German) Society of Paediatric Oncology (GPO)—SIOP II. Med Pediatr Oncol 25:166-178, 1995
Deutsch M, Thomas PR, Krischer J, et al: Results of a prospective randomized trial comparing standard dose neuraxis irradiation (3,600 cGy/20) with reduced neuraxis irradiation (2,340 cGy/13) in patients with low-stage medulloblastoma: A combined Children's Cancer Group-Pediatric Oncology Group Study. Pediatr Neurosurg 24:167-177, 1996
Thomas PR, Deutsch M, Kepner JL, et al: Low-stage medulloblastoma: Final analysis of trial comparing standard-dose with reduced-dose neuraxis irradiation. J Clin Oncol 18:3004-3011, 2000
Mulhern RK, Kepner JL, Thomas PR, et al: Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced-dose craniospinal irradiation: A Pediatric Oncology Group study. J Clin Oncol 16:1723-1728, 1998
Packer RJ, Goldwein J, Nicholson HS, et al: Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: A Children's Cancer Group Study. J Clin Oncol 17:2127-2136, 1999
Ris MD, Packer R, Goldwein J, et al: Intellectual outcome after reduced-dose radiation therapy plus adjuvant chemotherapy for medulloblastoma: A Children's Cancer Group study. J Clin Oncol 19:3470-3476, 2001
Grill J, Renaux VK, Bulteau C, et al: Long-term intellectual outcome in children with posterior fossa tumors according to radiation doses and volumes. Int J Radiat Oncol Biol Phys 45:137-145, 1999
Pendergrass TW, Milstein JM, Geyer JR, et al: Eight drugs in one day chemotherapy for brain tumors: Experience in 107 children and rationale for preradiation chemotherapy. J Clin Oncol 5:1221-1231, 1987
Chastagner P, Olive D, Philip T, et al: Efficacy of the "8 drugs in a day" protocol in brain tumors in children. Arch Fr Pediatr 45:249-254, 1988
Gentet JC, Bouffet E, Doz F, et al: Preirradiation chemotherapy including "eight drugs in 1 day" regimen and high-dose methotrexate in childhood medulloblastoma: Results of the M7 French Cooperative Study. J Neurosurg 82:608-614, 1995
Gentet JC, Doz F, Bouffet E, et al: Carboplatin and VP 16 in medulloblastoma: A phase II study of the French Society of Pediatric Oncology (SFOP). Med Pediatr Oncol 23:422-427, 1994
Feeny D, Furlong W, Boyle M, et al: Multi-attribute health status classification systems: Health Utilities Index. Pharmacoeconomics 7:490-502, 1995
Le Gales C, Costet N, Gentet JC, et al: Cross-cultural adaptation of a health status classification system in children with cancer: First results of the French adaptation of the Health Utilities Index Marks 2 and 3. Int J Cancer Suppl 12:112-118, 1999
Carrie C, Hoffstetter S, Gomez F, 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 45:435-439, 1999
Eberhart CG, Burger PC: Anaplasia and grading in medulloblastomas. Brain Pathol 13:376-385, 2003
Tait DM, Thornton-Jones H, Bloom HJ, et al: Adjuvant chemotherapy for medulloblastoma: The first multi-centre control trial of the International Society of Paediatric Oncology (SIOP I). Eur J Cancer 26:464-469, 1990
Evans AE, Jenkin RD, Sposto R, et al: The treatment of medulloblastoma: Results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine, and prednisone. J Neurosurg 72:572-582, 1990
Taylor RE, Bailey CC, Robinson K, et al: Results of a randomized study of preradiation chemotherapy versus radiotherapy alone for nonmetastatic medulloblastoma: The International Society of Paediatric Oncology/United Kingdom Children's Cancer Study Group PNET-3 Study. J Clin Oncol 21:1581-1591, 2003
Zeltzer PM, Boyett JM, Finlay JL, 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 17:832-845, 1999
Thiesse P, Jaspan T, Couanet D, et al: A protocol for imaging pediatric brain tumors. J Radiol 82:11-16, 2001
Fouladi M, Gajjar A, Boyett JM, et al: Comparison of CSF cytology and spinal magnetic resonance imaging in the detection of leptomeningeal disease in pediatric medulloblastoma or primitive neuroectodermal tumor. J Clin Oncol 17:3234-3237, 1999
Kortmann 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 46:269-279, 2000
Brand WN, Schneider PA, Tokars RP: Long-term results of a pilot study of low dose cranial-spinal irradiation for cerebellar medulloblastoma. Int J Radiat Oncol Biol Phys 13:1641-1645, 1987
Halberg FE, Wara WM, Fippin LF, et al: Low-dose craniospinal radiation therapy for medulloblastoma. Int J Radiat Oncol Biol Phys 20:651-654, 1991
Levin VA, Rodriguez LA, Edwards MS, et al: Treatment of medulloblastoma with procarbazine, hydroxyurea, and reduced radiation doses to whole brain and spine. J Neurosurg 68:383-387, 1988
Hughes EN, Shillito J, Sallan SE, et al: Medulloblastoma at the Joint Center for Radiation Therapy between 1968 and 1984: The influence of radiation dose on the patterns of failure and survival. Cancer 61:1992-1998, 1988
Packer RJ, Gajjar R, Vezina G, et al: Preliminary results of a prospective randomized study of "reduced" craniospinal radiotherapy followed by one of two adjuvant chemotherapy regimens for children with medulloblastoma: Children's Oncology Group study. Ann Neurol 56:s89-s90, 2004 (suppl 8)
Mulhern RK: Correlation of the Health Utilities Index Mark 2 cognition scale and neuropsychological functioning among survivors of childhood medulloblastoma. Int J Cancer Suppl 12:91-94, 1999
Palmer SL, Goloubeva O, Reddick WE, et al: Patterns of intellectual development among survivors of pediatric medulloblastoma: A longitudinal analysis. J Clin Oncol 19:2302-2308, 2001
Gajjar A, Hernan R, Kocak M, et al: Clinical, histopathologic, and molecular markers of prognosis: Toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 22:984-993, 2004
Fernandez-Teijeiro A, Betensky RA, Sturla LM, et al: Combining gene expression profiles and clinical parameters for risk stratification in medulloblastomas. J Clin Oncol 22:994-998, 2004(V. Oyharcabal-Bourden, C.)