Journal of Neuro-Oncology

, Volume 108, Issue 1, pp 163–171

Long-term results of combined preradiation chemotherapy and age-tailored radiotherapy doses for childhood medulloblastoma

Authors

    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Graziella Cefalo
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Daria Riva
    • Development Neurology UnitsFondazione IRCCS, Istituto Neurologico Carlo Besta
  • Veronica Biassoni
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Filippo Spreafico
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Emilia Pecori
    • Radiotherapy UnitsFondazione IRCCS, Istituto Nazionale Tumori
  • Geraldina Poggi
    • Acquired Brain Lesions UnitIRCCS Eugenio Medea
  • Paola Collini
    • Pathology UnitsFondazione IRCCS, Istituto Nazionale Tumori
  • Bianca Pollo
    • Neuropathology UnitFondazione IRCCS, Istituto Neurologico Carlo Besta
  • Laura Valentini
    • Department of Neurosurgery IIFondazione IRCCS, Istituto Neurologico Carlo Besta
  • Paolo Potepan
    • Department of RadiodiagnosticsFondazione IRCCS, Istituto Nazionale Tumori
  • Ettore Seregni
    • Nuclear Medicine UnitsFondazione IRCCS, Istituto Nazionale Tumori
  • Michela Casanova
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Andrea Ferrari
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Roberto Luksch
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Daniela Polastri
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Monica Terenziani
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Federica Pallotti
    • Nuclear Medicine UnitsFondazione IRCCS, Istituto Nazionale Tumori
  • Carlo Alfredo Clerici
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Elisabetta Schiavello
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Fabio Simonetti
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Cristina Meazza
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Serena Catania
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Marta Podda
    • Pediatric Oncology UnitFondazione IRCCS, Istituto Nazionale Tumori
  • Lorenza Gandola
    • Radiotherapy UnitsFondazione IRCCS, Istituto Nazionale Tumori
Clinical Study

DOI: 10.1007/s11060-012-0822-7

Cite this article as:
Massimino, M., Cefalo, G., Riva, D. et al. J Neurooncol (2012) 108: 163. doi:10.1007/s11060-012-0822-7

Abstract

To reduce the sequelae of craniospinal irradiation (CSI) in children under 10 (≥3) years old and to improve the prognosis for high-risk medulloblastoma in adolescents, we adjusted postoperative chemotherapy and CSI doses to patients’ stage and age. From 1986 to 1995, 73 patients entered the study. Children under 10 and adolescents with metastases, residual disease (RD) or stage >T3 received postoperative IV vincristine and high-dose (HD) ± intrathecal (IT) methotrexate, while standard-risk adolescents were given IV vincristine and IT methotrexate. Chemotherapy was followed by CSI (19.8 Gy for children <10; 36 Gy for adolescents), with a 54-Gy posterior fossa boost. Maintenance chemotherapy with lomustine and vincristine was administered for a year afterwards. A total of 39 children were under 10 of whom 20 had metastases. Response to chemotherapy was recorded in 70%, but 5-year EFS and OS were only 48 and 56%, respectively. Results were significantly worse for metastatic cases, patients under 10, those with RD, and those staged without MRI (unavailable early in the study). Efforts to preserve survivors’ quality of life did not pay off, and most patients over 30 still depended on their parents’ income and had severe cognitive/endocrine disabilities. In conclusion, despite a very high response rate with this preradiation HD methotrexate schedule, the outcome for high-risk medulloblastoma patients did not improve (especially when lower CSI doses were used) and patients still developed severe morbidities.

Keywords

Childhood medulloblastomaReduced craniospinal radiotherapyHigh-dose methotrexateIntrathecal chemotherapy

Introduction

When the treatment strategy described here was designed, the best results for medulloblastoma were achieved with CSI using a total dose of 35–36 Gy plus a 54-Gy boost to the posterior fossa. The reported severe long-term sequelae [1, 2] demanded innovative therapeutic strategies.

Our study aimed to investigate whether preradiation chemotherapy could make a lower CSI dose suffice in children under 10 years old and improve the outcome in adolescents at high-risk of recurrence. It failed on both counts because children with metastatic and residual disease had a very poor prognosis, and patients surviving after markedly reduced CSI did not have a better quality of life. The introduction of magnetic resonance imaging (MRI) and improved surgical and radiation techniques makes it difficult to compare treatment strategies adopted 25 years ago with those described nowadays, but our findings seem worth reporting because few results are available on the long-term follow-up of patients given reduced doses of CSI.

Patients and methods

Between 1987 and 1994, we enrolled 73 children with medulloblastoma aged 3–21 years; all diagnoses were confirmed by institutional review.

Tumor extent was evaluated preoperatively by computerized tomography (CT) ± myelography and/or MRI. After surgery, brain CT and/or gadolinium-enhanced and lumbar cerebrospinal fluid (CSF) cytology were always performed. From 1990 onwards, all children underwent spinal MRI, generally after surgery. Patients were staged according to Chang’s system [3], and stages T3–T4 or residual disease >1.5 cm2 (RD) and/or metastases (M+) were considered at high-risk (like RD and/or metastases alone are today). Treatment was administered according to patients’ age, extent of surgical resection and tumor size/extent.

Treatment

Surgical resection or biopsy was performed in all patients, aiming for maximal safe resection, all with parents’ and/or patients’ consent.

Chemotherapy

Table 1 shows the pre- and post-CSI chemotherapy schedule for the two subgroups: patients under 10 and adolescents with M+, RD or stage >T3 versus standard-risk cases over 10 years old. Four intrathecal methotrexate doses (10 mg/sqm) before each IV HD-MTX infusion were prescribed for the first 35 patients enrolled in the first group, and subsequently omitted due to evidence of a higher neurological toxicity risk of intrathecal MTX [4]. CSF sampling 24 h after methotrexate infusions revealed concentrations exceeded the level of 1 cytotoxic for leukemia (median 1.32 microM/L, data not shown).
Table 1

Pre- and post-CSI chemotherapy treatment plans for patient sub-groups

https://static-content.springer.com/image/art%3A10.1007%2Fs11060-012-0822-7/MediaObjects/11060_2012_822_Tab1_HTML.gif

aOnly the first 35 pts

CT chemotherapy, S surgery, it intrathecal, po per os; ci continuos infusion, CF folic acid

In all cases, chemotherapy was followed by CSI, scheduled 3–4 weeks after completing the preradiation chemotherapy. Three to four weeks after CSI, maintenance chemotherapy consisting of 6 cycles of IV VCR (1.4 mg/sqm every 3 weeks) and oral lomustine (80 mg/sqm every 9 weeks) was administered to all patients for 1 year in all.

Radiotherapy

All patients were treated using 6 MeV photons. In children over 10, the total CSI dose was 36 Gy (1.8 Gy/fraction to the brain and 1.5 Gy/fraction to the spine, with 5 fractions/week with a 54-Gy posterior fossa boost. In younger children, the total CSI dose was 19.8 Gy. The posterior fossa’s volume extended from the C1–C2 interspace to 1 cm above the midpoint between the foramen magnum and the cranial vertex. The volume of the CSI included the whole brain, spinal cord, and theca, up to the inferior edge of S2. Field junctions were displaced by 1 cm to ensure a uniform dose to gaps between fields each time a dose of 10 Gy was reached. The face, eyes, and front of the neck were shielded with lead.

Assessing response

All children were re-staged before planning radiation therapy. CSF cytology was repeated if it was positive for malignant cells before starting chemotherapy. Responses were defined as: complete response (CR), the complete disappearance of all radiographic and/or cytological evidence of tumor; partial response (PR), a more than 50% reduction in the product of the perpendicular diameters of the tumor with negative cytology; stable disease (SD), a <25% reduction in tumor size; and progressive disease (PD), a more than 25% increase in tumor size or the appearance of tumor in previously uninvolved areas [5].

Statistical considerations

Patient follow-up was brought up to date as at February 2011. The event-free survival rate (EFS) and overall survival rate (OS) were calculated from the start of chemotherapy to progression/relapse, second tumor or death (whichever came first) for EFS, and to death (due to any cause) for OS. The Kaplan–Meier estimator [6] was used to estimate EFS and OS, the log-rank test to compare the outcome of different patient groups, and the χ2 test to compare the frequency of patients’ characteristics. A P value <0.05 was considered statistically significant.

Neurodevelopmental assessment, growth and endocrine surveillance

Follow-up assessments of neurofunctional status, including acoustic sensitivity, linear growth, endocrine function, and hormone replacement needs, were planned in advance for children to deal with growth and thyroid abnormalities (i.e. growth hormone and gonadotropin tests, bone metabolism, thyroid function, and parenchymal alterations, etc.).

Details on the neuropsychological assessment of patients given intrathecal MTX are reported elsewhere [7].

Results

Patients

The male-to-female ratio was 1.9. The median age was 9.5 years (range 3–21); 39 children were 3–10 years old. Patients were staged at diagnosis using CT ± myelography in 18 cases and MRI ± gadolinium in the other 55. Tumor extent was T2 in 15 cases, T3 in 51, and T4 in 7. Table 2 groups patients by age, presence/absence of RD after surgery (NED/RD), and presence/absence of metastases (M+/M0). Nodular metastases were found by CT/myelography in 2 cases, by MRI in 5; only the CSF was positive for malignant cells in 13 cases.
Table 2

Presence/absence of residual disease and metastases in patients grouped by age

 

NED

ED

TOT

3–10 years

 M0

19

8

27

 M+

3

9

12

 Total

22

17

39

>10 years ‘high-risk’

 M0

9

8

17

 M+

3

5

8

 Total

12

13

25

>10 years ‘standard-risk’

 M0

9

0

9

Patients with tumor evidence after surgery are indicated in bold

NED no evidence of disease after surgery, ED evidence of disease after surgery, M0 no metastatic disease, M+ metastatic disease (M1, M2, M3)

Treatment

Forty-seven of the 64 eligible children completed preradiation HD-MTX treatment, while it was discontinued in 3 after the third cycle due to suspected progression (confirmed by CT and/or CSF cytology), and in 14 after 1–3 cycles due to delayed methotrexate clearance (toxic levels were identified after 24 h). All 9 standard-risk patients over 10 years old completed the planned preradiation treatment with IV vincristine plus IT methotrexate.

The median interval between surgery and irradiation was 107 days (range 74–142) for the regimen including HD-MTX, and 65 days (range 49–75) for the lower-risk regimen.

Response to HD methotrexate plus vincristine

Thirty-six of the 64 patients given preradiation VCR and HD-MTX had measurable disease in the posterior fossa (n = 30) and/or metastatic deposits (n = 20), postoperatively. Three children’s responses were not evaluated because they proceeded to radiotherapy after only one chemotherapy course. Combining findings on response at the primary site, spine, and CSF, these 33 patients had 13 CR, 10 PR, 7 SD, and 3 PD. The objective response rate (CR + PR) was therefore 70%. In all, 10/13 patients with M1 disease had CSF remission.

The disease progressed while on chemotherapy, after two courses of HD-dose MTX, in 2 of 37 children with no evidence of disease after surgery (M0NED).

Event-free and overall survival

The median follow-up was 185 months (range 104–268). The 5- and 10-year survival rates were, respectively, 56 ± 6% and 46 ± 6% for OS, and 48 ± 6% and 42.5 ± 6% for EFS.

Considering all patients, the presence of metastases coincided with a 5-year EFS and OS of 20 ± 10% and 25 ± 10%, respectively, for the 20 children whose tumors spread; for the 53 without metastases, the 5-year EFS and OS were 58 ± 7% (P = 0.0003) and 68 ± 6% (P < 0.0001), respectively (Fig. 1). The use of MRI for initial staging proved to be a significant prognostic variable, the 5-year EFS being 53 ± 7% for children staged with MRI as opposed to 33 ± 11% (P = 0.05) for those staged with CT ± myelography, and the 5-year OS was 60 ± 7% and 50 ± 12% (P = 0.02), respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-012-0822-7/MediaObjects/11060_2012_822_Fig1_HTML.gif
Fig. 1

EFS and OS according to presence/absence of metastases

Children 3–10 years old

Overall, the 5-year EFS and OS were, respectively, 38 ± 8% and 41 ± 8% (Fig. 2). The 5-year EFS was 17 ± 11% and 46 ± 10%, respectively, for patients with and without metastatic deposits (P = 0.0038). The 5-year EFS was 58 ± 11% for the 19 children without RD and 25 ± 15% for the 8 with RD; the OS was 58 ± 11% and 37 ± 7%, respectively (P = 0.04). For the 19 M0NED children, the 5-year EFS and OS were 58 ± 11% and 78 ± 11%, respectively; the 20 high-risk patients had a 5-year EFS and OS of 20 ± 9% (P = 0.04) and 25 ± 10%, respectively (P = 0.0023).
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-012-0822-7/MediaObjects/11060_2012_822_Fig2_HTML.gif
Fig. 2

EFS and OS according to age in patients receiving pre-RT HD-MTX

Children >10 years old

The 5-year EFS and OS were 59 ± 8% and 73 ± 8%, respectively. The 5-year EFS for the 25 high-risk patients was 64 ± 10%, while for the 9 low-risk cases, it was 44 ± 17% (P = 0.36). The OS was 69 ± 12% in the former and 78 ± 10% in the latter (P = 0.66).

The 5-year EFS was 61 ± 11% for the 18 cases without RD and 87 ± 12% for the 8 whose resection was incomplete (P = 0.11); their 5-year OS was 78 ± 10% and 100%, respectively (P = 0.16).

For the 26 non-metastatic patients, the EFS and OS were 69 ± 9% and 85 ± 7%, respectively; for the 8 cases with metastases, they were 25 ± 15% (P = 0.01) and 37.5 ± 17%, respectively (P = 0.002).

Role of chemotherapy, response and compliance

Intrathecal methotrexate influenced neither the EFS nor the OS.

Twenty-three patients achieving CR + PR had a 5-year EFS and OS of 43 ± 10% and 56 ± 10%, respectively, whereas 10 patients with less satisfactory responses (SD + PD) had rates of 10 ± 9% for both EFS (P = 0.0023) and OS (P = 0.0097). Finally, when the 57 patients who completed the schedule were compared with the 14 unable to do so, the 5-year EFS and OS did not differ.

Relapses

Forty-three patients had an adverse event, i.e. progression or relapse in 41 instances, and secondary tumors in 2. Progression or relapse became apparent within a median 22 months (range 1–93) of starting chemotherapy: 3 patients relapsed in the posterior fossa alone, 34 in the neuraxis but not locally, and 3 at both sites. One child relapsed in the albuginea testis through a ventricular shunt, with synchronous CSF dissemination. There were 93% disseminated relapses in all. The median time from relapse to death was 14 months; only one girl relapsing at a single spinal site was alive after 11 years at the time of this report.

Two patients treated with low-dose CSI developed secondary glioblastomas in supratentorial areas 120 and 288 months after diagnosis; both died.

Present-day functional outcome

Of the 31 long-term survivors, 15 are still being actively followed up, while updates on the others were obtained with brief phone calls: the details are given in Table 3.
Table 3

Follow-up history of long-term survivors (n = 31)

UPN/sex

Age at diagnosis

OS (months)

CSI doses

It MTX

Age at last contact

Engagement/marriage

Progeny

TS/F

5

144

Low

y

28

No

-

RL/M

4

161

Low

y

28

No

-

OV/M

7

135

Low

y

29

No

-

MS/M

4

97

Low

 

23

No

No

ME/M

9

110

Low

 

29

No

-

MA/F

8

113

Low

 

28

No (previous engagement)

No

LS/F

8.3

96

Low

 

27

No

No

FG/M

7

104

Low

 

26

No

No

CM/M

7.6

133

Low

y

29

No

No

CC/F

8

92

Low

 

26

No

No

AS/M

3.3

96

Low

 

31

No

No

AF/F

4

133

Low

y

26

Previous engagement

No

TGM/M

20

164

High

 

43

No

-

RM/F

10

82

High

 

30

No

-

RA/M

16

53

High

 

31

No

-

PG/M

14

154

High

y

37

No

-

PF/M

15

133

High

y

37

No

-

PA/M

12

168

High

y

37

No

-

MS/M

12

160

High

y

36

No

-

MS/F

13

101

High

 

32

No

-

ME/F

12

144

High

y

35

No

No

MD/F

12

161

High

y

36

Yes

1

MA/M

13

158

High

y

37

No

-

ILF/F

13

145

High

y

35

No

No

GA/M

11

85

High

 

28

No

No

DAV/F

14.7

81

High

y

32

Engaged

No

CG/M

13

133

High

y

35

Engaged

No

CD/M

14.5

116

High

y

36

No

No

CA/F

21

104

High

y

40

No

No

BMC/F

18

99

High

y

37

Divorced

1

AE/F

15

127

High

 

34

No

No

UPN/sex

Education

Housing

Employment

Endocrine therapy

Other remarks

MRI alterations

Secondary tumors

TS/F

Professional school

Parents

No

NO

Severe mental delay

Leukoencephalopathy

No

RL/M

Secondary school: economics

Parents

Hospital secretariat

T4, GH

Driving licence, dysarthria

Leukoencephalopathy

No

OV/M

Secondary school: technique sciences

Parents

Private technical employer

T4

Hypercholesterolemia

No

Thyroid papillary carcinoma

MS/M

Secondary school: foreign languages

Parents

Restaurant clerk

T4

Driving license, many friends

No

No

ME/M

Special school for disables

Parents

None

GH, T4

Previously disable

Severe leukoencephalopathy

Meningioma, progressing

MA/F

Primary school

Parents

Housekeeper

T4, short stature

Bilateral auditive deficits, small stature

Gliosis

No

LS/F

University degree

Parents

No

T4, short stature

Nucal alopecia, cheratitis

Meningioma

Meningioma

FG/M

Secondary school: classic literature

Parents

None

No

Hyatal hernia

No

No

CM/M

Secondary school: technique sciences

Parents

Tessil firm worker

T4

Nucal alopecia, short stature

Mild leukoencephalopathy

No

CC/F

Professional school

Parents

Artigianal worker

T4, GH, estroprogestinic

Antidepressive

2 CR

No

AS/M

Primary school

Parents

None

T4, GH, obesity

Epylepsia

No

Multiple thyroid nodules

AF/F

Nursery teacher

Parents

Bakery worker

T4, short height

Nucal alopecia

No

No

TGM/M

Uncomplete secondary school

Parents

insurance employer

T4, statins, antihypertensive

Ataxia

Leukoencephalopathy

Secondary thyroid papillary carcinoma

RM/F

Secondary school

Parents

White collar

No

 

Meningiomas

Meningiomas

RA/M

Secondary school

Parents

Television operator

T4, GH

 

No

No

PG/M

Unknown

Miss

Worker

No

 

No

No

PF/M

Secondary school

Parents

White collar

T4

 

No

Thyroid nodule

PA/M

Professional school

Parents

Lost 2 years ago

T4

 

No

No

MS/M

Primary school

Parents

None

GH

Anti-epileptics for 4 years

 

Meningioma

MS/F

University degree

Parents

None

No

  

Meningioma

ME/F

Secondary school: teacher

Parents

Tessil firm worker

T4

Ataxia, few activities and friends, driving license

Severe leukoencephalopathy

No

MD/F

Primary school

Husband

Housekeeper

No

Unknown

Unknown

Unknown

MA/M

Special school for disables

Parents

Artigianal worker

GH

Previously disable

Severe leukoencephalopathy

 

ILF/F

Primary school

Parents

White collar

T4

Benzodiazepine ataxia, nucal alopecia

Leukoencephalopathy

No

GA/M

University degree

Parents

None

GH, T4

Severe ataxia

Leukoencephalopathy

Struma

DAV/F

University degree

Parents

None

T4

On psycotherapy

Left frontal hemosiderin deposits

No

CG/M

Professional school

Parents

Private artigianal worker

 

Ant-acid for hyatal hernia

Leukoencephalopathy

No

CD/M

Professional school

Parents

Blind category

T4

Amaurosis

Periventricular gliosis

No

CA/F

University degree

Parents

University teacher

T4, GH

Nucal alopecia, short stature

No

No

BMC/F

Uncompleted secondary school

With the son

None

T4

Ataxia, left deft, osteoporosis, cheratitis, familiar polyposis

Left parietal hemosiderin deposits

No

AE/F

Secondary school for teacher

Parents

None

T4, cm 160

Mild dysmetria

No

No

Only 6 long-term survivors have had any sentimental relationships, two have had children, two have their own home. Although 17 have had a formal education beyond primary school, only 11 have a stable job. The vast majority of patients are hypothyroid and report stable or worsening neurological deficits with cosmetic problems relating to alopecia.

Discussion

Over the last 30 years, medulloblastoma has been the subject of a number of studies undertaken to improve the disease’s prognosis by combining chemotherapy with postoperative radiotherapy [8, 9].

Just how much benefit, in terms of more limited neuropsychological sequelae, might derive from CSI dose reduction is not entirely clear [10, 11]. Goldwein et al. [10] reported that reduced CSI (18 Gy) associated with intensive postradiation chemotherapy was followed by a largely normal intellectual development in 7/10 survivors. A contemporary study by Jakacki et al. [11] on 7 patients was stopped because, although none of them progressed while on preradiation chemotherapy (4 months of cisplatin/carboplatin, vincristine, cyclophosphamide and etoposide), 3 patients relapsed and none of the survivors were without endocrine deficits and schooling problems. In the randomized POG [12] (Pediatric Oncology Group) study, a reduction in CSI from 36 to 23.4 Gy, without additional chemotherapy, resulted in significantly better intellectual abilities [13] in the only 22/35 survivors accessible; after 8 years, the EFS was 60% for children given standard-dose radiotherapy and 52% for those given a reduced dose (P = 0.141) [14]. There was a high rate of extraneural relapses in the POG study, suggesting that chemotherapy might help to reduce the relapse rate at systemic sites.

Packer et al. [15] reported excellent survival rates using reduced CSI doses (2,340 cGy) followed by adjuvant chemotherapy with VCR, lomustine, and cisplatin, confirming these results in the largest randomized trail published to date on medulloblastoma. The results were excellent, with 5-year EFS and OS both over 80%.

The decision to use HD-MTX in our strategy was based on evidence that this drug could cross the blood–brain barrier, and on the therapeutic responses reported in relapsing patients [16].

Two randomized studies of the 1990s (the CCG 921 and the German HIT’91) compared sandwich chemotherapy with immediate radiotherapy followed by adjuvant CT, finding the latter more beneficial in both trials [17, 18]. The SIOP II randomized reduced-dose CSI, with or without preradiation chemotherapy, in children with non-disseminated medulloblastoma [19]. The 5-year EFS, with or without chemotherapy, for children given standard-dose CSI (3,500 cGy) was 67.6% as opposed to 55.3% for those given a reduced-dose (2,500 cGy) (P = 0.07). Children given preradiation chemotherapy followed by reduced-dose CSI fared poorly (5-year EFS 41%). The subsequent multicenter randomized PNET-3 study was the only one to describe a better EFS for preradiation chemotherapy than radiotherapy alone for nonmetastatic medulloblastoma [20].

We conducted a phase 2 study on preradiation chemotherapy with vincristine plus HD methotrexate, obtaining a satisfactory 70% response rate, but the survival rates were not as expected in children under 10 years. The results in non-metastatic patients were negatively influenced by the fact that this group included patients with unidentified metastases (because MRI was not available for the staging of nearly 25% of patients).

Children over 10 treated with HD-MTX and full-dose irradiation had a more satisfactory 5-year EFS over 60% and 5-year OS of 78 ± 10%, which were comparable with other protocols applying radiation therapy alone.

The danger of reducing CSI doses and delaying radiation without providing sufficiently intensive chemotherapy is also underscored by the fact that 93% of the relapses encountered were disseminated. Even full compliance with the chemotherapy regimen and the addition of intrathecal methotrexate in some children failed to improve their prognosis. Intrathecal MTX did not have the positive impact reported by Rutkowski et al. [21], who obtained excellent results in low-risk children with medulloblastoma without any irradiation. Part of their success was probably due to a more modern, consistent, and refined staging, and to their more intensive regimen including multiple non-cross-reactive drugs.

As for the goal of containing or avoiding damage to patients’ cognitive performance, we have already reported on a subgroup of 21 long-term survivors, in which patients under 10 fared much worse than their siblings if they had received intrathecal methotrexate in addition to CSI, even in low doses [7].

Looking at the latest status of long-term survivors, there is very little cause for satisfaction: a very small percentage of patients are socially or affectively independent, although 10 of them received low-dose CSI. Similar data have been reported in children receiving a total dose of 18 Gy for CNS prophylaxis in leukemia without suffering any brain injury [22]. The role of systemic HD methotrexate in generating severe and symptomatic leukoencephalopathy seems more doubtful, however, as other authors have already stated [4, 23]. The combination of IV and IT doses of methotrexate with full CSI radiation doses should nonetheless be avoided. Final outcome may have also been strongly influenced by duration of hydrocephalus, socio-economic status, school attendance, and other aspects not considered when patients were assessed. Had it been available at the time, the intensive care protocol that we now adopt (providing physiatric, auxological and psychological support that was unavailable in the 1980s) might also have positively influenced patient outcome.

Only intensive chemotherapy regimens can control the disease and enable risk-related dose stratification even in high-risk patients [9], while 4 HD methotrexate courses can induce a very high response rate but are not curative, especially in cases of local and disseminated disease with reduced CSI doses.

Acknowledgments

This work was partially supported by Associazione Bianca Garavaglia onlus, Busto Arsizio, Italy, provided resident scholarships

Conflict of interest

None.

Copyright information

© Springer Science+Business Media, LLC. 2012