1688 / 1708
Results
Table
1shows patient demographic information and
treatment characteristics, and Table
2shows control de-
mographic information. The mean age at cranial irradi-
ation was 11.8 years (range 1.1–18.6). The primary site
was infratentorial in 5
/
19, supratentorial in 12
/
19, and
leukemia in 2
/
19. Radiation treatment plans were cra-
niospinal,
n
¼
8; whole brain radiation,
n
¼
3; and 3-
dimensional or intensity-modulated RT,
n
¼
9. Mean
prescription dose was 42.9 Gy (range 12 Gy–59.4 Gy).
Eight patients (42%) received concurrent chemotherapy,
most commonly (87.5%) vincristine based. The neuro-
psychological evaluation was completed by the follow-
ing number of patients: baseline,
n
¼
13 (time between
diagnosis and baseline testing: range 14–2284 days,
median 127 days); 6-month follow-up,
n
¼
13;
15-month follow-up,
n
¼
14; 27-month follow-up,
n
¼
10; and by the following number of controls: baseline,
n
¼
55; 6-month follow-up,
n
¼
43; 15-month
follow-up,
n
¼
38; 27-month follow-up,
n
¼
37.
Group Differences in Neuropsychological Performance
Fig.
2shows overall changes in neuropsychological test
scores over time in the patient and control groups, and
Fig.
3shows the differences in the mean test scores at in-
dividual visits. The main LME analysis indicated that
patients tended to have a significantly lower (worse)
overall performance on motor dexterity (group,
P
¼
.015; Group
×
Time,
P
¼
.027), verbal learning
(group,
P
¼
.001; Group
×
Age
0
,
P
¼
.003), visuospa-
tial working memory (group–trend,
P
¼
.057;
Group
×
Age
0
,
P
¼
.047; Group
×
Time,
P
¼
.003),
and visual perception (group,
P
,
.0001; Group
×
Age
0
,
P
,
.0001; Group
×
Time,
P
¼
.018; Group
×
Age
0
×
Time,
P
¼
.015). Patients had poorer neuropsy-
chological performance than controls as early as base-
line, with significantly lower test scores on motor
dexterity (Purdue Pegboard,
P
¼
.008) and verbal learn-
ing (Memory for Words,
P
¼
.003) (Fig.
3 ). Performance
improved with age in both groups on motor dexterity
(Purdue Pegboard, Fig.
2 A), verbal learning (Memory
for Words, Fig.
2B), visuospatial working memory
(Bead Memory, Fig.
2C), and visual perception (visual
perception test, Fig.
2D) (all tests,
P
,
.0001).
However, different rates of change in test performance
were found with age, such that neuropsychological def-
icits relative to controls—in particular verbal learning,
visuospatial working memory, and visual perception—
were more pronounced in younger patients (Fig.
2 ). In
Table 1.
Patient and treatment characteristics
n
%
Gender
Male
12
63
Female
7
37
Age at diagnosis
0–4 y
4
21
5–9 y
3
16
10–14 y
8
42
15–19 y
4
21
Ethnicity
Caucasian
13
68
African American
4
21
Other
2
11
Handedness
Right
18
95
Left
1
5
Diagnosis
Glioma
4
21
Medulloblastoma
/
PNET
5
26
Germinoma
3
16
Leukemia
2
11
Nongerminoma germ cell tumor
2
11
Pineoblastoma
1
5
Craniopharyngioma
1
5
Ependymoma
1
5
RT technique
Craniospinal
8
42
Whole brain
3
16
3D conformal or IMRT
8
42
Radiation dose
0–20 Gy
2
11
21–49 Gy
3
16
50–60 Gy
14
74
Primary site
Supratentorial
14
74
Infratentorial
5
26
Concurrent chemotherapy
Yes
8
42
No
11
58
Abbreviations: PNET, primitive neuroectodermal tumor; IMRT,
intensity-modulated RT.
Table 2.
Control characteristics
n
%
Gender
Male
30
55
Female
25
45
Age at enrollment
0–4 y
0
0
5–9 y
19
35
10–14 y
21
38
15–19 y
15
27
Ethnicity
Caucasian
26
47
African American
24
44
Other
5
9
Handedness
Right
44
80
Left
11
20
Redmond et al.: Radiation to neural progenitor niches and neurocognitive outcomes
NEURO-ONCOLOGY
†
M A R C H 2 0 1 3
363
at Universitaet Leipzig, Institut fuer Informatik/URZ, Bibliothek on August 25, 2014
http://neuro-oncology.oxfordjournals.org/Downloaded from