1520 / 1708
ratio was greater than unity for 6 of 9 patients who have dis-
ease control, with a median index of 1.4 (range, 0.1–7.2).
Two of the 3 patients with progressive disease experienced
failure at metastatic sites in the brain or spine that were ag-
gressively resected and treated with supplemental irradiation
to 59.4 Gy after high-dose CSI.
For patients with metastatic disease treated by using CSI,
there was 1 case each of documented secondary malignancy,
necrosis, and myelopathy. The patient with the longest PFS
in this subgroup developed a histologically confirmed high-
grade secondary glial neoplasm that arose 5 years after irra-
diation to a site of metastatic disease in the supratentorial
brain. Total dose to this site was 59.4 Gy. The patient with
the shortest interval from initiation of RT
1
to initiation of
RT
2
, 4 months, developed necrosis at the site of previous fo-
cal treatment that was included in the CSI volume. This pa-
tient received focal irradiation to the fourth ventricle after
GTR and experienced progression with metastatic disease
at L
3
–S
1
only 10 weeks after completing his first treatment
course. After resection, CSI (39.6 Gy) with boost treatment
of L
3
–S
1
(59.4 Gy) was delivered. Necrosis occurred in the
cerebellum approximately 6 months after completing RT
2
and subsequently was treated by using resection and
HBOT. The region of necrosis received approximately 99
Gy. The patient remains with neurologic deficits and no evi-
dence of progressive disease 2 years after completion of sal-
vage therapy. The other case was a patient with a history of
supratentorial EP treated by using previous focal irradiation.
This patient experienced failure with metastases to Meckel’s
cave, the cervical spine, and conus medullaris. The patient
underwent metastasectomy of the intracranial metastases
and cervical spinal cord metastasis. After GTR, this patient
received CSI to 39.6 Gy and focal treatment to all known sites
of metastases, including 54 Gy to the cervical spinal cord.
The patient was seen in follow-up for her 2-year evaluation
after initiation of RT
2
and was noted to have bilateral
lower-extremity weakness. Magnetic resonance imaging of
the spinal cord showed extensive T2 changes at the level of
previous resection. The HBOT was administered, and 1
year later, the patient returned to her asymptomatic baseline
and imaging changes resolved. The PFS for subsets of re-
ported patients is shown in
Fig. 4 .Median combined total
dose at any point of overlap in the brain was 99 Gy (range,
86.4–117 Gy).
Overall survival for three subsets of reported patients is
shown in
Fig. 5 .None of the 12 patients with initial meta-
static failure treated by using CSI died despite progression
in 3 patients. There were three deaths in the 13 patients
with initial local failure treated with FFRT. The overall sur-
vival estimate for this subset was 67% 16% at 5 years.
There was only one survivor in the patients treated with ra-
diosurgery after local failure. The 5-year overall survival es-
timate was 20% 18%.
DISCUSSION
Children with EP tend to be young, with more than 50% of
cases diagnosed in patients younger than 3 years. Children in
this age range are most vulnerable to the effects of RT, and its
use in frontline management is both recent and experimental.
Historically, the pattern of failure for children with EP treated
with surgery and RT was local. With improving rates of GTR
and image-guided high-dose irradiation to 59.4 Gy, the pat-
tern of failure has become mixed, with a greater proportion
of patients experiencing failure with metastatic disease
(5) .The combination of young age and metastatic failure are
two prominent factors that drive investigators to test
72
48
24
0
Time (months)
1.0
0.8
0.6
0.4
0.2
0.0
Initial Treatment Failure
(1)
(2)
(3)
Fig. 3. Initial rates and patterns of failure after surgery and radiation
therapy. Combined local and metastatic failure (black–(1)), local
failure (red–(3)), and metastastic failure (blue–(2)).
144
120
96
72
48
24
0
Time (months)
1.0
0.8
0.6
0.4
0.2
0.0
Progression-Free Survival
(1)
(2)
(3)
Fig. 4. Progression-free survival after reirradiation according to
treatment method and initial tumor pattern failure (blue (1) = 12 pa-
tients with metastatic failure treated with craniospinal reirradiation;
green (2) = 13 patients with local failure retreated with focal frac-
tionated irradiation; red (3) = 5 patients with local failure treated
with radiosurgery).
94
I. J. Radiation Oncology
d
Biology
d
Physics
Volume 71, Number 1, 2008