High-dose Chemotherapy With Autologous Stem CellRescue Followed by Posterior Fossa Irradiation forLocal Medulloblastoma Recurrence or ProgressionAfter Conventional Chemotherapy

Vita Ridola, MD1

Jacques Grill, MD, PhD1

Francois Doz, MD, PhD2

Jean-Claude Gentet, MD3

Didier Frappaz, MD4

Marie-Anne Raquin, MD1

Jean-Louis Habrand, MD, PhD5

Christian Sainte-Rose, MD6

Dominique Valteau-Couanet, MD, PhD1

Chantal Kalifa, MD1

1 Department of Pediatric and Adolescent Oncol-ogy, Gustave Roussy Institute, Villejuif, France.

2 Department of Pediatric Oncology, Curie Insti-tute, Paris, France.

3 Department of Pediatric Oncology, UniversityHospital ‘‘La Timone,’’ Marseille, France.

4 Department of Pediatric Oncology, Leon BerardCenter, Lyon, France.

5 Department of Radiation Oncology, GustaveRoussy Institute, Villejuif, France.

6 Department of Neurosurgery, Necker-Sick Chil-dren Hospital, Paris, France.

BACKGROUND. The objective of the current study was to determine the outcome of

children with local recurrence or progression of medulloblastoma in patients who

received high-dose chemotherapy (HDC) and posterior fossa (PF) irradiation.

METHODS. HDC consisted in busulfan at a dose of 600 mg/m2 and thiotepa at a

dose of 900 mg/m2 followed by autologous stem cells transplantation (ASCT). PF

radiotherapy was delivered at doses from 50 grays (Gy) to 55 Gy on Day 170 after

ASCT. Twenty-seven patients developed local recurrence of an initially completely

resected medulloblastoma. Twelve patients had local residual disease after surgery

and were enrolled into the salvage protocol at the time of local disease progression

under conventional chemotherapy.

RESULTS. Acute toxicity consisted mainly in hepatic veno-occlusive disease (33% of

patients) and bone marrow aplasia. Two toxic deaths (5%) from infections were

reported. The 5-year overall survival rate after this salvage treatment (OS5y) for the

39 children who were treated was 68.8% (95% confidence interval [95% CI], 53–

81.2%). In the group of patients who were treated for local recurrence, the OS5y was

77.2% (95% CI, 58.3–89.1%). Patients with local residual disease who were treated at

the time of disease progression had an OS5y after salvage treatment of only 50%

(95% CI, 25.4–74.6%; P 5 .09).

CONCLUSIONS. The treatment strategy that was used in this study had manageable

immediate toxicity and resulted in a high overall survival rate in the setting of young

children with medulloblastoma who developed local recurrence or disease progres-

sion. Cancer 2007;110:156–63.� 2007 American Cancer Society.

KEYWORDS: primitive neuroectodermal tumors, recurrence, salvage therapy,busulfan, thiotepa, intellectual outcome.

M edulloblastoma is one of the most common malignant brain

tumors in children. Despite gradual improvements in the sur-

vival rate over the last decades, infants and young children still have

a dismal prognosis with poor overall survival (OS) and high treat-

ment-related morbidity.1 Greater attention has been given to long-

term sequelae (with special concerns regarding neurocognitive

impairment) as the treatment for medulloblastoma has become more

effective. Although a variety of factors play a role in the development

of such sequelae, it has been demonstrated that age at diagnosis, per-

ioperative complications, and craniospinal irradiation (CSI) are the

major causes of this impairment.2–5

To decrease long-term sequelae caused by central nervous sys-

tem irradiation in young children with medulloblastoma, most

Presented at the 11th Congress of the Interna-tional Society of Pediatric Neuro-Oncology, Bos-ton, Massachusetts, June 13�16, 2004, and atthe 36th Congress of the International Society ofPediatric Oncology, Vancouver, British Columbia,Canada, September 21�24, 2005.

Address for reprints: Jacques Grill, MD, PhD,Department of Pediatric and Adolescent Oncol-ogy, Institut Gustave Roussy, 39, rue CamilleDesmoulins, 94805 Villejuif, France; Fax: (011)33-1-42115275; E-mail: grill@igr.fr

Received December 12, 2006; revision receivedFebruary 25, 2007; accepted March 6, 2007.

ª 2007 American Cancer SocietyDOI 10.1002/cncr.22761Published online 31 May 2007 in Wiley InterScience (www.interscience.wiley.com).

156

cooperative groups have designed strategies based on

prolonged conventional chemotherapy to delay or

avoid radiotherapy.6–8 In patients with recurrent dis-

ease, salvage strategy usually includes CSI with poste-

rior fossa (PF) boost; however, radiotherapy still puts

a serious strain on the neurocognitive outcome,

because most patients are young.9,10 In 1990, the

French Society of Pediatric Oncology (SFOP) Brain Tu-

mor Committee elaborated both an initial strategy,

which was based on conventional adjuvant chemo-

therapy without radiation (the Baby Brain SFOP

[BBSFOP] protocol), and a salvage treatment, which

was based on high-dose chemotherapy (HDC) fol-

lowed by autologous stem cell rescue (ASCR) and

involved-field radiotherapy.11 The HDC regimen con-

sisted of high-dose busulfan and thiotepa, because

encouraging results were reported from a Phase II

study of children with refractory, malignant brain

tumors12 and also because of the effectiveness of the

regimen in avoiding CSI.13 In this report, we present

the results from this salvage HDC regimen associated

with local-field radiotherapy in children with local

disease recurrence/progression of medulloblastoma.

MATERIALS AND METHODSPatient SelectionChildren with standard-risk medulloblastoma who

were diagnosed before age 5 years were treated with

conventional adjuvant chemotherapy, without irra-

diation, according to the BBSFOP protocol.11 Salvage

strategies were indicated in the protocol according to

the results from the initial pilot studies.12,13

Children who presented with local failure, either

during or after BBSFOP chemotherapy, were consid-

ered eligible for salvage treatment with high-dose

busulfan and thiotepa and PF irradiation (PFRT).

Local failure was defined as radiologic reappearance

or progression of disease at the original site. An in-

termediate analysis of the BBSFOP study in 1996 in-

dicated that conventional chemotherapy was unable

to control overt residual disease; thus, the protocol

was amended, and patients then received HDC and

PFRT as initial treatment (results to be reported else-

where). Informed consent for the salvage procedure

was obtained from the parents separately at the ini-

tial treatment with conventional chemotherapy. Chil-

dren who had metastatic recurrences in craniospinal

fluid and/or on magnetic resonance imaging (MRI)

studies were excluded from this study and received

a sequential HDC and craniospinal radiotherapy regi-

men.11 The diagnosis of medulloblastoma was

confirmed after a central pathology review in all

patients.

Treatment RegimenSalvage therapy consisted of oral busulfan at a dose

of 150 mg/m2 per day (37.5 mg/m2 every 6 hours)

for 4 consecutive days on Days �8, �7, �6, and �5,

and thiotepa at a dose of 300 mg/m2 per day for the

next 3 days on Days �4, �3, and �2, administered as

a 1-hour intravenous infusion. Autologous stem cells

(either bone marrow or peripheral stem cells) were

reinfused 48 hours after the completion of chemo-

therapy, on Day 0. During the 7 days of chemother-

apy, hyperhydration (3 L/m2 per day) was

administered, and all patients received clonazepam

during chemotherapy to prevent busulfan-related sei-

zures. Ursodeoxycholic acid was given orally at a

dose of 300 mg/m2 per day from Day �9 to Day 180

to reduce the risk of hepatic veno-occlusive disease

(VOD).14 All patients received granulocyte colony-sti-

mulating factor at a dose of 5 mcg/kg per day start-

ing 5 days after ASCR until their neutrophil count

recovered to >0.5 3 109/L. Platelet transfusions were

administered to maintain the platelet count at

>50 3 109/L, and packed red cell transfusions were

given to maintain hemoglobin levels at >7 g/dL. Feb-

rile neutropenia was treated with broad-spectrum

antibiotics and antifungal or antiviral agents, as

appropriate. Parenteral nutrition was given when

necessary. All patients were hospitalized in the trans-

plantation unit in single rooms with laminar air-flow.

Starting in 1996, as soon as they entered the sal-

vage protocol and before they received HDC, chil-

dren received 2 courses of etoposide at a dose of 100

mg/m2 per day as a 1-hour infusion from Day 1 to

Day 5 and carboplatin at a dose of 160 mg/m2 per

day as a 1-hour infusion also from Day 1 to Day 5.

Combined chemotherapy with etoposide and carbo-

platin was adopted for peripheral stem cell mobiliza-

tion.

SurgeryMaximal surgical removal was recommended when-

ever possible. It usually was performed in patients

with residual disease after HDC and before radiother-

apy. In some cases, children underwent surgery

before they received HDC, either at the time they

entered the salvage protocol or after 2 courses of eto-

poside plus carboplatin.

Resection was considered total if the neurosur-

geon recorded total resection with no recognizable

residual disease or tumor adhesions both in the sur-

gical report and on postoperative imaging studies

that were negative for macroscopic residual disease,

in accordance with guidelines from the International

Society of Pediatric Oncology.15 If patients had radi-

ologic residual disease or incomplete resection on

HDC for Pediatric Medulloblastoma/Ridola et al. 157

the surgical report, then resection was considered

less than total.

RadiotherapyRadiotherapy was delivered only to the PF after at least

70 days after the day of autologous stem cell transplan-

tation (ASCT). Initial radiotherapy with 2 opposed lat-

eral beams was adopted to treat the entire PF at doses

from 50 Gy to 55 Gy in 1.8-Gy daily fractions over

6 weeks. CSI at 35 Gy was indicated only in patients

with progressive disease after HDC. In the most recent

years, since 1997, children received conformal PF RT

at a dose of 35 Gy with an additional boost of 15 Gy

restricted to the tumor bed.

Toxicity CriteriaToxicity criteria were graded according to the

National Cancer Institute Common Toxicity Criteria

(version 3.0).16 Liver toxicity was graded according to

the Bearman toxicity criteria.17

Tumor ResponseTumor response was assessed by MRI studies that

were obtained 6 weeks after HDC and before the

start of radiotherapy. Children who received �2

courses of standard chemotherapy also were evalu-

ated by MRI studies before the administration of

HDC. Response was defined according to published

criteria for tumor response in children with brain

tumors.18

Radiologic and Neuropsychological Follow-upAfter the completion of the treatment, children

underwent clinical examinations and MRI neuroima-

ging every 3 months during the first 2 years and ev-

ery 6 to 12 months thereafter. Neurodevelopmental

tests were scheduled to be administered between 6

months to 1 year after radiotherapy and annually

thereafter. Children were tested with age-adapted

Wechsler scales.19,20

Statistical AnalysisOS rates were estimated using the Kaplan-Meier

method21 and were calculated from the date of radi-

ologic diagnosis of recurrence or progression to the

date of death or last follow-up visit for patients who

remained alive. Progression-free survival (PFS) rates

were estimated using the above-mentioned method

from the date of radiologic recurrence or progression

to the date of either documented failure (date of

radiologic or clinical recurrence or death) or the last

follow-up visit for patients in complete remission

(CR). The 95% confidence intervals (95% CIs) for sur-

vival rates were estimated using the Rothman

method.22 Follow-up data were updated in February

2007.

RESULTSHistory Before HDCBetween January 1988 and July 2005, 39 children in 8

French pediatric oncology departments were enrolled

in the study. The median age at diagnosis was

31 months (range, 8–58 months), and the median

age at the time patients were started on the salvage

protocol was 39 months (range, 11–80 months).

All children had previously received the conven-

tional BBSFOP chemotherapy. Twenty-seven children

who underwent complete surgical removal developed

local disease recurrence during or after BBSFOP

chemotherapy (median delay of recurrence, 8 months;

range, 1–17 months), and 12 children experienced

the progression of local surgical residual disease dur-

ing chemotherapy (median delay of progression,

5 months; range, 1–14 months).

None of these 39 children presented with distant

metastases at the time of enrolment or had evidence of

distant metastases in craniospinal fluid samples or on

craniospinal MRI studies. At the initial histopathologic

examination, 27 children had classic medulloblas-

toma, 11 children had desmoplastic medulloblastoma,

and 1 child had medullomyoblastoma.

Twenty-five children received other additional

treatments prior to HDC. Nine children underwent

surgery, 7 of whom underwent radiologically com-

plete removal of the tumor. Seventeen children

received 2 courses of etoposide plus carboplatin to

allow stem cell collection while treating the disease

with the most effective conventional chemotherapy

combination23: Two of those patients achieved CR

with this chemotherapy regimen, 9 patients achieved

a partial response (PR), 3 patients had stable disease,

and 3 patients were not evaluable because of previ-

ous complete surgical removal of their tumor. Four

children received just 1 course of etoposide plus car-

boplatin (2 PRs and 2 responses that were not evalu-

ated), and 2 children received 1 and 4 courses of

other chemotherapies (etoposide plus cyclophospha-

mide and high-dose ifosfamide, respectively). In

total, 13 of 18 cases of disease recurrence/progres-

sion (72%) responded to the combination of etopo-

side and carboplatin. At the time HDC was initiated,

30 children had measurable disease, and 9 children

were in CR because of surgery or chemotherapy.

Response to HDCTwenty-nine patients who had measurable disease

remained alive and were evaluable for response at

158 CANCER July 1, 2007 / Volume 110 / Number 1

the end of HDC. Among them, 9 patients achieved a

CR, and 9 patients achieved a PR, resulting in an

objective response rate of 62%. Seven children had

stable disease, and 4 patients had progressive dis-

ease.

Surgery After HDCNine of 20 children who had residual disease after

HDC underwent surgical removal. In 5 of those chil-

dren, a macroscopic total resection was performed,

and the other 4 children underwent a less than mac-

roscopic total resection. In 2 patients, no tumor cells

were found. In 7 of 9 children, histologic examina-

tion of the residual tumor showed the presence of

viable medulloblastoma cells; however, in 2 of those

children, the pathologist identified just a few isolated

medulloblastoma cells inside a fully fibrotic tissue.

Acute Toxicity After TransplantationAcute toxicity of the high-dose busulfan plus thio-

tepa regimen was similar to the toxicity described

previously by our group with this regimen.12,13,24 The

median duration of neutropenia <0.5 3 109/L was 9

days (range, 6–21 days), and the median duration of

thrombocytopenia <50 3 109/L was 30 days (range,

5–120 days).

Severe infectious complications occurred in 11

children, including 7 patients who presented with

bacterial septicemia, 3 patients with focal pneumo-

nias, and 1 patient with pulmonary aspergillosis. Two

toxic deaths were observed among 39 children

(5.1%), both of which were caused by infectious

complications: one case was related to Staphylococ-

cus aureus septicemia with acute respiratory distress

syndrome, and 1 case was related to pulmonary

aspergillosis.

Hepatic VOD occurred in 13 of 39 children

(33.3%). In 5 of 39 children (12.8%), grade 3 VOD

was observed. Clinical recovery was possible in all

patients with symptomatic treatment, not including

defibrotide. There was no correlation noted between

the incidence or severity of VOD and previous chem-

otherapy.

ASCT consisted of bone marrow in 12 children,

peripheral stem cells in 26 children, and both in 1

child. The median duration of hospitalization in the

transplantation unit was 40 days (range, 24–97 days).

RadiotherapyAmong the 39 children who were included in the

current study, 37 children completed radiotherapy.

One child died during HDC, and 1 child died during

radiotherapy, both from infectious complications.

Among the 37 irradiated children, 4 patients received

craniospinal irradiation in addition to the PF boost

because of progressive disease during chemotherapy.

One patient was treated with g-knife radiation (a

major protocol violation), and he experienced a local

recurrence a few months later and died of tumor

progression. Thirty-two patients received local radio-

therapy on the PF only at a dose of 50 grays (Gy) to

54 Gy in 1.8-Gy daily fractions. Twenty-two patients

were irradiated on the entire PF; whereas, after 1997,

10 children received conformal radiotherapy at a

dose of 35 Gy on the PF with a boost of 15 Gy to the

tumor bed. The 4 children who received CSI received

radiotherapy to the whole PF.

Outcome and Treatment FailureFor the entire cohort of 39 children, the 5-year OS

and event-free survival (EFS) estimates were 68.8%

(95% CI, 53–81.2%) and 61.5% (95% CI, 45.9–75.1%),

respectively (Fig. 1). EFS and OS for both patient

subgroups are shown in Figures 2 and 3.

Patients who developed a local recurrence of a

previous completely removed medulloblastoma had

a 5-year OS estimate of 77.2% (95% CI, 58.3–89.1%)

and a 5-year EFS estimate of 66.7% (95% CI, 47.8–

81.4%). Patients who developed local progression of

postsurgical radiologic residual disease during

BBSFOP chemotherapy had a 5-year OS estimate of

50% (95% CI, 25.4–74.6%) and a 5-year EFS estimate

of 50% (95% CI, 25.4–74.6%).

Treatment failures occurred in 13 patients within

the first 2 years of salvage therapy, and the median

delay of disease recurrence was 10 months (range,

FIGURE 1. Overall and event-free survival of children who received thecombination of busulfan and thiotepa regimen. Survival curves were updated

as of February 2007 for all surviving patients using the Kaplan-Meier

method. The median follow-up of the cohort was 4.5 years (1 patient died of

a late neurologic complication with aspiration pneumonia). HDC indicates

high-dose chemotherapy.

HDC for Pediatric Medulloblastoma/Ridola et al. 159

7–24 months). Patterns of recurrence were as follows:

5 local recurrences, 4 metastatic disseminations, and

4 synchronous local and metastatic recurrences. One

patient who developed a PF recurrence among the

10 patients who received reduced PF irradiation, and

3 of 23 patients underwent whole PF irradiation

(Fisher exact test; P value not significant).

Among the 37 evaluable patients, disease recur-

rences were observed in 1 of 18 patients who were in

CR after HDC and in 12 of 19 patients who were not

in CR after HDC (including patients who entered CR

after further surgery or radiotherapy; (chi-square test;

P 5 .0002). The patterns of failure did not differ

depending on the 2 different therapeutic situations.

Neuroradiologic Abnormalities at Follow-upIn 9 of 39 patients (23%), a peculiar phenomenon of

brain injury was observed on follow-up imaging stu-

dies (Fig. 4). Brain lesions that were characterized by

abnormal contrast enhancement on MRI studies,

with a punctiform aspect or a giriform distribution,

and without mass effect, appeared with a median

delay of 6 months (range, 1–18 months) after the

completion of radiotherapy. The localization of these

lesions could be cortical or subcortical, but in all

cases they appeared only inside the irradiation field.

The majority of the children (6 of 9 patients) were

asymptomatic at onset, whereas 3 children presented

with a worsening of their cerebellar deficit. These

lesions evolved spontaneously or after corticosteroids

toward regression in 8 of 9 patients (88%) after 6 to

12 months with the appearance of calcifications. Pre-

transplantation treatment did not appear to have an

impact on the occurrence of this complication.

Neurocognitive DevelopmentAt the most recent evaluation, among the 20 children

who had at least 1 Intellectual Quotient (IQ) exami-

nation after salvage treatment at a median follow-up

of 5.8 years after diagnosis, the mean estimated full-

scale IQ (FSIQ), verbal IQ, and performance IQ

values were 71.8 (range, 40–100), 79.3 (range, 45–

110), and 70.6 (range, 45–101), respectively. The me-

dian age at the last evaluation was 7.5 years (range,

4.7–15.2 years). The FSIQ tended to decrease over

time at a mean rate of 2 points per year from the

date of diagnosis (Fig. 5).

No statistically significant difference was

observed in the mean FSIQ of 69.3 in 9 of 20 chil-

dren with who had radiologic abnormalities at fol-

low-up compared with the mean FSIQ of 73.8 in 11

of 20 children who had normal MRI studies (Student

t test; P value not significant). Among the 12 children

who were aged <3 years at the time of diagnosis, the

mean FSIQ at a mean follow-up of 70 months was

70.6 (range, 50–94) whereas, among the 8 children

who were aged >3 years at diagnosis, the mean FSIQ

at a mean follow-up of 43 months was 73.6 (range,

40–100; Student t test; P value not significant).

FIGURE 2. Event-free survival of children who received the combination ofbusulfan and thiotepa (BU-THIO) according to the circumstances of their sal-

vage therapy (Kaplan-Meier method). Patients received conventional chemo-

therapy according to Baby Brain French Society of Pediatric Oncology

(BBSFOP) protocols either after they developed a local recurrence (n 5 27

patients) or after the progression of a local residual tumor (n 5 12 patients).

Differences in the overall survival rates were not found to be significant

(P 5 .28; log-rank test). HDC indicates high-dose chemotherapy; gr, grade.

FIGURE 3. Overall survival of children who received the combination ofbusulfan and thiotepa (BU-THIO) according to the circumstances of their sal-

vage therapy (Kaplan-Meier method). Patients (pts) received conventional

chemotherapy according to Baby Brain French Society of Pediatric Oncology

(BBSFOP) protocols either after a local recurrence (n 5 27 patients) or after

the progression of a local residual tumor (n 5 12 patients). Differences in

the overall survival rates were of borderline statistical significance (P 5 .09;

log-rank test). HDC indicates high-dose chemotherapy; gr, grade.

160 CANCER July 1, 2007 / Volume 110 / Number 1

DISCUSSIONYoung children with medulloblastoma who develop

local recurrence or disease progression after conven-

tional chemotherapy can be treated effectively with

the combined strategy of high-dose busulfan plus

thiotepa and radiotherapy to the PF, as demonstrated

in the current study by the observed high 5-year OS

rate of 68.8%. To our knowledge, there is no compa-

rable series in the literature. Standard CSI-based sal-

vage therapies yielded a 5-year OS rate of 37% in 13

patients aged <4 years who had M0 medulloblas-

toma and received treatment with conventional

chemotherapy only at the time of diagnosis25; there

is no indication in this publication that the efficacy

of the salvage strategy depended on the type of dis-

ease recurrence or progression. It is worth noting

that, in our strategy, we used only focal irradiation

and therefore induced fewer long-term neurologic

side effects.

Acute toxicity was manageable but characterized

mainly by a significant percentage of hepatic VOD in

33% of patients. None of the children died of liver

toxicity. This liver toxicity reportedly is caused

mainly by busulfan but can be modified by the

accompanying drug.26 It is likely that the rate and se-

verity of VOD could be decreased by refinement of

the use of busulfan (ie, an intravenous form of

busulfan27) or by the prophylactic use of defibro-

tide.28

Delayed toxicities from the treatment strategy

included mainly the gonads and the central nervous

system. Major toxicity includes the risk of irreversible

ovarian failure resulting from the use of high-dose

busulfan, as reported by Teinturier et al, who

observed a 100% rate of severe and persistent ovarian

failure in 10 female patients with a median follow-up

of 7 years who were treated during childhood with

FIGURE 4. Example of a major radiochemotherapeutic interaction after treatment combining busulfan and thiotepa and whole posterior fossa irradiation at adose of 50 grays. Treatment was administered for local recurrence in a child aged 28 months. Nine months after the completion of treatment, her ataxia wor-

sened together with the appearance of abnormal gadolinium enhancement in the radiation field. Using steroid therapy only, her neurologic situation improved

and the radiologic images gradually decreased.

FIGURE 5. Longitudinal evaluation of children who received the salvageregimen. All Development Quotients (Brunet-Lezine) and Intellectual Quotients

(IQs) (age-adapted Wechsler scales) since the diagnosis of the disease are

plotted in the figure. Depending on the duration of follow-up, 27 children

had between 1 and 6 evaluations (median, 3 evaluations). The equation of

the tendency line shows a decline of 1.9 points of IQ per year and an inter-

cept of 83.6 points (estimated mean initial IQ of the whole cohort). Children

who could not be evaluated were located in centers distant from the neurop-

sychology teams. FSIQ indicates full-scale IQ.

HDC for Pediatric Medulloblastoma/Ridola et al. 161

high-dose busulphan-containing regimens.29 Cryo-

preservation of ovarian tissue is an option to address

the sterility issue; although, currently, it is only at the

experimental level.30,31

Neurotoxicity has been observed as a delayed

side effect characterized by transient radiologic

anomalies in the irradiated fields in 23% of patients.

The pathophysiologic hypothesis of this phenom-

enon is in favor of blood-brain barrier damage sec-

ondary to disruption of the endothelium of small

vessels induced by either radiotherapy, HDC, or

(most likely) a combination of both treatments being

‘‘guilty by association.’’32 Indeed, busulfan alone

has no delayed neurologic toxicity,33 and PF irradia-

tion only at the doses used in the current study

rarely has been associated with radionecrosis. Busul-

fan is a known radiosensitizer with documented skin

reactions.34 Calcifications are the latest consequence

of this microangiopathic phenomena and have

been described previously in other radiochemo-

therapic interactions, such as those observed with

methotrexate.

The cognitive outcome associated with the cur-

rent strategy, with a median FSIQ of 71.8 at 5.8 years

of follow-up, is an improvement on previously

reported results after CSI-based salvage therapy (me-

dian FSIQ, 62 at 4.8 years of follow-up in children

aged <4 years at diagnosis).25 Despite the positive din IQ values between the 2 different salvage proto-

cols, long-term intelligence outcomes still were unsa-

tisfactory. To enhance the quality of life of long-term

survivors, future clinical research should focus on

improving the cognitive outcome. Possible strategies

may be represented by the adoption of new radio-

therapy techniques (ie, proton radiotherapy, reduced

boost volumes, or the use of hyperfractionated radio-

therapy). Changing the type of drugs in the HDC

regimen may not lower the risk of radiochemothera-

peutic interactions, because these phenomena al-

ready have been described with thiotepa35 and

cyclophosphamide.36

In conclusion, the results from this multicenter

study indicated that the combination of busulfan

and thiotepa with PF irradiation is an efficient sal-

vage regimen in children with medulloblastoma who

develop recurrent or progressive local disease. Cur-

rently, we are evaluating whether this regimen would

be more useful for patients with medulloblastoma

who have local residual disease if it was administered

at the time of diagnosis. Its immediate toxicity is

manageable and may be improved with more speci-

fic treatments of VOD; however, delayed neurologic

and gynecologic side effects remain main concerns

and warrant treatment refinements.

REFERENCES1. Kuhl J, Doz F, Taylor RE. Embryonic tumors. In: Walker DA,

Perilongo G, Punt JAG, Taylor RE, eds. Brain and Spinal Tumors

of Childhood. London: Arnold Publisher; 2004:314–330.

2. Mulhern RK, Palmer SL, Reddick WE, et al. Risk of young

age for selected neurocognitive deficits in medulloblas-

toma are associated with white matter loss. J Clin Oncol.

2001;19:472–479.

3. Suc E, Kalifa C, Brauner R, et al. Brain tumors under the

age of three. The price of survival. Acta Neurochir. 1990;

106:93–98.

4. Palmer SL, Goloubeva O, Reddick WE, et al. Patterns of in-

tellectual development among survivors of pediatric medul-

loblastoma: a longitudinal analysis. J Clin Oncol. 2001;19:

2302–2308.

5. Grill J, Viguier D, Kieffer V, et al. Critical risk factors for in-

tellectual impairment in children with posterior fossa

tumors: the role of cerebellar damage. J Neurosurg. 2004;

101:152–158.

6. Duffner PK, Horowitz ME, Krischer JP, et al. Postoperative

chemotherapy and delayed radiation in children less than

three years of age with malignant brain tumors. N Engl J

Med. 1993;17:1725–1731.

7. Geyer JR, Zeltzer PM, Boyett JM, et al. Survival of infants

with primitive neuroectodermal tumors or malignant epen-

dymomas of the CNS treated with eight drugs in 1 day: a

report from the Childrens Cancer Group. J Clin Oncol. 1994;

12:1607–1615.

8. Geyer JR, Sposto R, Jennings M, et al. Multiagent chemo-

therapy and deferred radiotherapy in infants with malig-

nant brain tumors: a report from the Children’s Cancer

Group. J Clin Oncol. 2005;23:7621–7631.

9. Gajjar A, Mulhern RK, Heideman RL, et al. Medulloblas-

toma in very young children: outcome of definitive cra-

niospinal irradiation following incomplete response to

chemotherapy. J Clin Oncol. 1994;12:1212–1216.

10. Fouladi M, Gilger E, Kocak M, et al. Intellectual and func-

tional outcome of children 3 years old or younger who

have CNS malignancies. J Clin Oncol. 2005;23:7152–7160.

11. Grill J, Sainte-Rose C, Jouvet A, et al. Treatment of medul-

loblastoma with postoperative chemotherapy alone: an

SFOP prospective trial in young children. Lancet Oncol.

2005;6:573–580.

12. Kalifa C, Hartmann O, Demeocq F, et al. High dose busul-

fan and thiotepa with autologous bone marrow transplan-

tation in childhood malignant brain tumors: a phase II

study. Bone Marrow Transplant. 1992;9:227–233.

13. Dupuis-Girod S, Hartmann O, Benhamou E, et al. Will

high-dose chemotherapy followed by autologous bone

marrow transplantation supplant cranio-spinal irradiation

in young children treated for medulloblastoma? J Neuroon-

col. 1996;27:87–98.

14. Essell JH, Schroeder MT, Harman GS, et al. Ursodiol pro-

phylaxis against hepatic complications of allogenic bone

marrow transplantation. A randomized, double-blind, pla-

cebo-controlled trial. Ann Intern Med. 1998;128:975–981.

15. Gnekow A. Recommendations of the Brain Tumor Subcom-

mittee for the reporting of trials. SIOP Brain Tumor Sub-

committee.International Society of Pediatric Oncology. Med

Pediatr Oncol. 1995;24:104–108.

16. National Cancer Institute. Common Terminology Criteria

for Adverse Events: version 3.0 (CTCAE). Available at URL:

http://ctep.cancer.gov/forms/CTCAEv3.pdf Accessed August

9, 2006.

162 CANCER July 1, 2007 / Volume 110 / Number 1

17. Bearman SI, Appelbaum FR, Buckner CD, et al. Regimen-

related toxicity in patients undergoing bone marrow trans-

plantation. J Clin Oncol. 1988;6:1562–1568.

18. Zeltzer PM, Friedman HS, Norris DG, Ragab AH. Criteria

and definitions for response and relapse in children with

brain tumors. Cancer. 1985;56:1824–1826.

19. Wechsler D. Manual for the Wechsler Preschool and Pri-

mary Scale of Intelligence. Revised ed. New York, NY: Psy-

chological Corporation; 1989.

20. Wechsler D. Wechsler Intelligence Scale for Children. 3rd ed.

New York, NY: Psychological Corporation; 1991.

21. Kaplan EL, Meier P. Nonparametric estimation from in-

complete observations. J Am Stat Assoc. 1958;53:457–448.

22. Rothman KJ. Estimation of confidence limits for the cumu-

lative probability of survival in life table analysis. J Chronic

Dis. 1978;31:557–560.

23. 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. 1994;23:

422–427.

24. Valteau-Couanet D, Filippini B, Grill J, et al. High dose

busulphan and thiotepa followed by autologous stem cell

transplantation (ASCT) in previously irradiated medullo-

blastoma patients: high toxicity and lack of efficacy. Bone

Marrow Transplant. 2005;36:939–945.

25. Walter AW, Mulhern RK, Gajjar A, et al. Survival and neuro-

developmental outcome of young children with medullo-

blastoma at St Jude Children’s Research Hospital. J Clin

Oncol. 1999;17:3720–3728.

26. Bouligand J, Boland I, Valteau-Couanet D, et al. In children

and adolescents, the pharmacodynamics of high-dose

busulfan is dependent on the second alkylating agent used

in the combined regimen (melphalan or thiotepa). Bone

Marrow Transplant. 2003;32:979–986.

27. Lee JH, Choi SJ, Lee JH, et al. Decreased incidence of

hepatic veno-occlusive disease and fewer hemostatic derange-

ments associated with intravenous busulfan vs oral busul-

fan in adults conditioned with busulfan 1 cyclophos-

phamide for allogeneic bone marrow transplantation. Ann

Hematol. 2005;84:321–330.

28. Richardson PG, Murakami C, Jin Z, et al. Multi-institutional

use of defibrotide in 88 patients after stem cell transplan-

tation with severe veno-occlusive disease and multisystem

organ failure: response without significant toxicity in a

high-risk population and factors predictive of outcome.

Blood. 2002;100:4337–4343.

29. Teinturier C, Hartmann O, Valteau-Couanet D, et al. Ovar-

ian function after autologous bone marrow transplantation

in childhood: high dose busulfan is a major cause of ovar-

ian failure. Bone Marrow Transplant. 1988;22:989–994.

30. Wallace WH, Anderson RA, Irvine DS. Fertility preservation

for young patients with cancer: who is at risk and what

can be offered? Lancet Oncol. 2005;6:209–218.

31. Poirot C, Brugieres L, Genestie C, Martelli H. Ovarian tissue

cryopreservation for prepubertal girls: indications and fea-

sibility. Gynecol Obstet Fertil. 2005;33:799–803.

32. Kieffer-Renaux V, Bulteau C, Grill J, et al. Visual agnosia af-

ter treatment of a posterior fossa ependymoma in a 16-

month-old girl. J Child Neurol. 2001;16:698–704.

33. Notteghem P, Soler C, Dellatolas G, et al. Neuropsychologi-

cal outcome in long-term survivors of a childhood extra-

cranial solid tumor who have undergone autologous bone

marrow transplantation. Bone Marrow Transplant. 2003;31:

599–606.

34. Vassal G, Hartmann O, Habrand JL, Pico JL, Lemerle J. Radio-

sensitization after busulphan [letter]. Lancet. 1987;1:571.

35. Cairncross G, Swinnen L, Bayer R, et al. Myeloablative

chemotherapy for recurrent aggressive oligodendroglioma.

Neuro-oncol. 2000;2:114–119.

36. Fouladi M, Chintagumpala M, Laningham FH, et al. White

matter lesions detected by magnetic resonance imaging af-

ter radiotherapy and high-dose chemotherapy in children

with medulloblastoma or primitive neuroectodermal tu-

mor. J Clin Oncol. 2004;22:4551–4560.

HDC for Pediatric Medulloblastoma/Ridola et al. 163