paediatrics Brussels 17

complete. Another patient considered to have undergone par-

tial resection underwent placement of a ventriculoperitoneal

shunt. Severe acute complications developed in 6 patients

consisting of cranial nerve paralysis with swallowing diffi-

culties in 3, akinetic mutism in 1, both disturbances in 1,

and hemiplegia in association with an intracranial postopera-

tive hematoma in 1.

Central radiologic review

Both postoperative CT and MRI were performed in 11

cases, MRI only in 11 cases, and CT only in 2. At the central

radiologic review, the resection was considered complete in

16 of the 24 patients and doubtful or partial in 4 each.

Of the 20 patients with infratentorial EP and the 4 patients

with supratentorial EP, 14 and 2, respectively, underwent

CR. In 6 children with infratentorial tumors, a discrepancy

was found between the operative report and the early postop-

erative imaging results. The central radiologic review

showed radiologic residual disease in 1 case that had been

considered a CR by the surgeon. In contrast, no residual

mass was found in 5 patients for whom the neurosurgeon

had reported subtotal surgery.

Only the extent of resection as defined by the panel of

experts was considered for statistical analysis.

Radiotherapy

Radiotherapy was performed in all patients, a median of 31

days (range, 20–97) after surgery. In 5 cases, RT was started

>40 days after surgery. In 1 case, the interval was as long as

97 days because of second-look surgery. That patient was

considered to have undergone a CR before the initiation of

RT. For all patients, the mean interval between RT sessions

was

6 h. The energy was >8 MeV in 22 cases and equal

to 6 MeV in 2. Of the 24 children, 18 received 60 Gy. Of

these 18 patients, 14 had undergone a CR, 1 an incomplete,

and 3 a doubtful resection, as determined after central review.

Another 5 children received 66 Gy. Of these 5, 1 had under-

gone a CR, 3 an incomplete, and 1 a doubtful resection. One

patient with a CR received 54 Gy owing to a protocol

violation. HFRT was conformational in 20 patients. Of the

4 patients with measurable residual disease, 3 had an objec-

tive response to HFRT. No severe acute complications of

HFRT were reported.

PFS and OS

The median follow-up was 87.5 months (95% confidence

interval, 66–90). Of the 24 patients, 8 died, all of neoplastic

evolution. The estimated 3- and 5-year OS rate was 79.2%

and 74.8%, respectively

( Fig. 1 )

. The 3- and 5-year PFS

rate was 62.5% and 54.2%, respectively.

Prognostic factors analysis

Tumor grade and extent of resection were selected to be

tested as potential prognosis factors. The extent of resection

was assessed by the central radiologic review committee. The

two-tailed test failed to find any statistically significant differ-

ence

( Table 1 )

Patterns of failure

At the last follow-up visit, 12 patients were in their first

complete remission, 2 were in their second or greater

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114

Months

PFS

3-year OS: 79.2

, IC95

[63.9; 95.4]; 3-year PFS: 62.5

, IC95

[43.1; 81.9]

5-year OS: 74.8

, IC95

[57.3; 92.3]; 5-year PFS: 54.2

, IC95

[34.2; 74.1]

Fig. 1. Overall survival (OS) and progression-free survival (PFS)

rates of 24 patients with intracranial ependymoma.

Table 1. Univariate analysis of correlation between selected parameters and estimated PFS and OS rates

Characteristic

5-y PFS (%)

5-y OS (%)

Grade

.849

.833

53.8 (24.8–76.0)

76.2 (42.7–91.7)

50.0 (18.3–75.3)

70.0 (25.3–82.7)

Location

.465

.854

Supratentorial

75.0 (12.8–96.1)

Infratentorial

50.0 (27.1–69.2)

74.7 (49.3–86.6)

Extent of resection

.842

.269

Complete

56.3 (29.5–76.2)

81.3 (52.4–93.5)

Partial or doubtful

50.0 (15.3–77.5)

62.5 (29.9–86.0)

Radiation dose (Gy)

.561

.991

52.6 (28.7–71.9)

73.7 (47.9–88.1)

80.0 (20.3–96.9)

Abbreviations:

PFS = progression-free survival; OS = overall survival.

Data in parentheses are 95% confidence intervals.

* As assessed by central radiologic review.

1538

I. J. Radiation Oncology

Biology

Physics

Volume 74, Number 5, 2009