Download Expression analysis of stem cell-related genes reveal OCT4 as a

J Neurooncol (2012) 106:71–79
DOI 10.1007/s11060-011-0647-9
LABORATORY INVESTIGATION - HUMAN/ANIMAL TISSUE
Expression analysis of stem cell-related genes reveal OCT4
as a predictor of poor clinical outcome in medulloblastoma
Carolina Oliveira Rodini • Daniela Emi Suzuki • Najsla Saba-Silva •
Andréa Cappellano • Jorge Estefano Santana de Souza • Sérgio Cavalheiro
Silvia Regina Caminada Toledo • Oswaldo Keith Okamoto
•
Received: 17 February 2011 / Accepted: 20 June 2011 / Published online: 2 July 2011
Ó Springer Science+Business Media, LLC. 2011
Abstract Aberrant expression of stem cell-related genes
in tumors may confer more primitive and aggressive traits
affecting clinical outcome. Here, we investigated expression and prognostic value of the neural stem cell marker
CD133, as well as of the pluripotency genes LIN28 and
OCT4 in 37 samples of pediatric medulloblastoma, the
most common and challenging type of embryonal tumor.
While most medulloblastoma samples expressed CD133
and LIN28, OCT4 expression was found to be more sporadic, with detectable levels occurring in 48% of tumors.
Expression levels of OCT4, but not CD133 or LIN28, were
significantly correlated with shorter survival (P B 0.0001).
Median survival time of patients with tumors hyperexpressing OCT4 and tumors displaying low/undetectable
OCT4 expression were 6 and 153 months, respectively.
More importantly, when patients were clinically stratified
according to their risk of tumor recurrence, positive OCT4
expression in primary tumor specimens could discriminate
patients classified as average risk but which further
deceased within 5 years of diagnosis (median survival time
of 28 months), a poor clinical outcome typical of high risk
patients. Our findings reveal a previously unknown prognostic value for OCT4 expression status in medulloblastoma, which might be used as a further indicator of poor
survival and aid postoperative treatment selection, with a
particular potential benefit for clinically average risk
patients.
Keywords Medulloblastoma CD133 LIN28 OCT4 Prognosis Stem cell markers
Electronic supplementary material The online version of this
article (doi:10.1007/s11060-011-0647-9) contains supplementary
material, which is available to authorized users.
C. O. Rodini D. E. Suzuki S. Cavalheiro
Department of Neurology and Neurosurgery, Federal University
of São Paulo, São Paulo, SP, Brazil
C. O. Rodini D. E. Suzuki O. K. Okamoto (&)
Centro de Estudos do Genoma Humano, Departamento de
Genética e Biologia Evolutiva, Instituto de Biociências,
Universidade de São Paulo, Rua do Matão 277 Cidade
Universitária, São Paulo, SP 05508-090, Brazil
e-mail: [email protected]
N. Saba-Silva A. Cappellano S. R. C. Toledo
Institute of Pediatric Oncology (GRAACC), Pediatric
Department, Federal University of São Paulo, São Paulo,
SP, Brazil
J. E. S. de Souza
Ludwig Institute for Cancer Research, São Paulo Branch
at Hospital Alemão Oswaldo Cruz, São Paulo, SP, Brazil
Introduction
Pediatric medulloblastoma is a malignant embryonal neuroepithelial tumor that accounts for approximately 16% of
all pediatric brain tumors, typically occurring in children
under the age 10. While treatable, about 15% of patients
die from the disease within 2 years of diagnosis [1, 2].
Current treatment is based mainly on tumor resection,
followed by craniospinal radiotherapy and treatment with
conventional cytotoxic agents. However, these complementary therapies often elicit long-term adverse effects
including secondary tumors, as well as hearing, cognitive,
endocrinal, and vascular impairment [3]. Thus, to manage
benefits and risks, the choice of postoperative treatment
modality is based on clinical parameters including age at
diagnosis, volume of the remaining tumor, and presence of
metastasis.
123
72
Children under 3 years of age are usually not subjected
to radiotherapy to avoid damage to the developing brain.
Older patients with non-metastatic and totally or nearly
totally resected tumors are considered as having ‘‘average
risk’’ of tumor recurrence and are treated with low doses of
radiation and chemotherapy. Patients with partial tumor
resection and/or metastatic disease are classified as ‘‘high
risk’’, regarding tumor recurrence, and thus subjected to a
more aggressive postoperative treatment with radiation and
chemotherapy at higher doses.
Under this rational treatment stratification based on
clinical prognostic criteria, the 5-year progression-free
survival rate for medulloblastoma stands at around 40 and
70% for high and average risk patients, respectively [4].
Still, a significant number of patients do not respond well
to standardized treatments. About 30% of average risk
patients either present disease progression or succumb to
the disease within 5 years. Although only a few alternative
experimental treatments are available for high risk
patients, an improved classification system could help to
identify beforehand those clinically average risk patients
with potential poor prognosis and select them to more
aggressive treatments, as performed for their high risk
counterparts.
Under the cancer stem cell hypothesis, factors contributing to poor prognosis, such as resistance to conventional
therapies, tumor recurrence, and metastasis, could be
associated with a small subset of highly tumorigenic stemlike cells. In medulloblastoma, stem-like cells were first
identified and isolated based on expression of CD133, a
cell membrane antigen present in neural stem cells [5]. As
an embryonal tumor, medulloblastoma cells could also
display ectopic expression of typical pluripotency genes
such as LIN28 and OCT4, which encode an RNA binding
protein and a transcription factor containing a POU
homeodomain, respectively [6]. In fact, OCT4 expression
has been reported in germ cell tumors including seminoma,
dysgerminoma, germinoma, and embryonal carcinoma [7],
and its expression has been associated with a more primitive and malignant tumor phenotype [8].
In medulloblastoma, expression of stem cell-related
genes might help distinguish tumors associated with distinct clinical outcomes, but, thus far, a possible correlation
of such stem cell markers with prognosis has not been
investigated in this important pediatric tumor. In this
study, we have analyzed the expression of CD133, LIN28,
and OCT4 genes in a series of 37 medulloblastoma samples and have found a significant correlation of OCT4
expression with overall survival. Interestingly, OCT4
expression could discriminate primary tumor specimens
from patients clinically classified as average risk but
which had a poor clinical outcome similar to that of high
risk patients.
123
J Neurooncol (2012) 106:71–79
Materials and methods
Tumor collection
A total of 37 medulloblastoma specimens were obtained by
surgical resection of patients admitted for diagnosis and
treatment at the Institute of Pediatric Oncology (IOP/
GRAACC), Federal University of São Paulo, Brazil, from
1990 to 2009. The study was approved by the Institutional
Ethics Committee (CEP/UNIFESP N8 CEP 1115/09) and
followed the criteria established by the Helsinki convention. Informed consent was obtained from all patients/
guardians.
All patients were classified according to the North
American Children’s Oncology Group trial into two
groups, based on age at initial diagnosis, resection extent,
and Chang metastasis staging [9]. Average risk patients
were those over 3 years of age at diagnosis, with nonmetastatic and totally or nearly totally resected tumors
(B1.5 cm2 on post-operative magnetic resonance imaging).
Patients not fulfilling these criteria were regarded as high
risk.
Gene expression analysis
Fresh surgical samples of primary medulloblastoma and
non-neoplastic brain tissues obtained from two patients
with Arnold Chiari Type I malformation were immediately
snap-frozen in liquid nitrogen upon surgical removal. Total
cellular RNA was extracted using TRIzolÒ Reagent
(Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions, and the RNA integrity was evaluated based on the intensity of 28S and 18S rRNA bands in
1% agarose gel. A total of 1 lg of RNA was reverse
transcribed with SuperScriptTM III Reverse Transcriptase Kit (Invitrogen) according to the manufacturer’s
instructions.
Expression levels of CD133 (NM_006017), OCT4
(NM_002701.4) and LIN28 (NM_024674.4) genes were
evaluated by quantitative real-time PCR. Amplification of
specific PCR products was detected with SYBR Green
(Power SYBRÒ Green PCR Master Mix kit; Applied
Biosystems, UK) according to the manufacturer’s protocol.
All qPCR reactions had a total volume of 25 ll, containing
1 ll of cDNA sample, 10 pmol of each primer and 12.5 ll
of SYBR Green. The thermal cycling was carried out by
starting with 95°C for 10 min hold, followed by 50
amplification cycles of 95°C for 15 s and 68°C for 30 s.
Reactions were performed in duplicate. Expression data
analysis was performed in a real-time PCR system
(MX3005PTM; Stratagene) and quantitation was based on
standard curves. Each run was completed with a melting curve analysis to confirm amplicon specificity.
J Neurooncol (2012) 106:71–79
The housekeeping gene encoding hypoxanthine guanine
phosphoribosyltransferase (HPRT, NM_000194.2) was used
as endogenous reference [10]. Normalized gene expression
was calculated as target gene/HPRT. Relative gene expression was estimated as normalized gene expression in tumor/
non-neoplastic control. Primer sequences were designed at
specific exon–exon junctions of the target gene to avoid DNA
amplification and sequence specificity was confirmed by a
BLAST search at the NCBI databank: OCT4 For: 50 -CCTGA
AGCAGAAGAGGATCACC-30 , OCT4 Rev: 50 -AAAGCG
GCAGATGGTCGTTTGG-30 ; LIN28 For: 50 -CCAGTGGA
TGTCTTTGTGCACC-30 , LIN28 Rev: 50 -GTGACACGGATG
GATTCCAGAC-30 ; HPRT For: 50 -CCTGGCGTCGTGAT
TAGTGA-30 , HPRT Rev: 50 -GCCTCCCATCTCCTTCATC-30 ; CD133 For: 50 -CACTACCAAGGACAAGGCG
TTC-30 , CD133 Rev: 50 -CAACGCCTCTTTGGTCTCC
TTG-30 .
Statistical analysis
Correlation of clinical variables (metastasis, risk groups,
surgical resection, relapse, and patient status) with gene
expression data was determined by the Mann–Whitney test.
Overall survival analysis from life-time data according to
clinical risk group, presence or absence of gene expression,
and relative gene expression levels (hyperexpression cutoff value of five-fold, relative to non-neoplastic control)
was determined by the Kaplan–Meier product-limit estimation with log-rank (P \ 0.05). Overall survival was
defined as time from diagnosis to the day of death or last
follow-up. The Spearman’s rank correlation coefficient and
the Fisher’s exact test to correlate numeric variables and
categorical data, respectively, were performed with the
software Statistical Package GraphPad Prism 5 (GraphPad
Software, CA, USA). For correlation of gene expression in
tumors, normalized expression (9107) were log transformed for graphical display. Significance was established
at the P B 0.05 level in all analyses.
Results
Clinicopathological parameters
Patients included in the study were mainly males (70%)
over 3 years old (81%), and tumors were primarily of the
classic histological subtype (89%). Based on age at initial
diagnosis, resection extent, and Chang metastasis staging,
17 patients (46%) were clinically stratified as average risk
and 20 patients (54%) as high risk (Table 1). As expected,
significant differences (P = 0.0271) in overall survival
were found according to clinical risk. The 5-year survival
rates for average and high risk patients were 70 and 33%,
73
respectively. Among high risk patients, 60% had metastasis
at diagnosis.
Expression of stem cell markers and prognosis
Expression of the neural stem cell marker CD133 could be
detected in most high (94%) and average risk (90%) tumor
samples. Similarly, the majority of high (85%) and average
(88%) risk tumor samples had detectable expression of the
pluripotency gene LIN28. In contrast, expression of another
pluripotency-related gene, OCT4, was more scattered
among tumors, being detected at low levels in 45 and 53%
of high and average risk tumor samples, respectively. When
analyzing the concurrent expression of these three genes in
tumor samples, significant positive correlation were found
only between LIN28 and OCT4 (P = 0.0014, Fig. 1a) and
between LIN28 and CD133 (P = 0.0052, Fig. 1b).
As shown in Fig. 2, detectable expression (presence or
absence) of OCT4, but not CD133 or LIN28, was significantly correlated with overall survival. Patients with
OCT4? tumors tend to have a shorter median survival time
(24 months) than those whose tumors were negative for
OCT4 expression (153 months). Such correlation was
found with tumors displaying any level of OCT4 expression, as detected by real-time PCR, and its level of significance was similar to that of clinical risk and survival
correlation.
Furthermore, when considering relative OCT4 expression levels, a stronger correlation (P \ 0.0001; Fig. 3a)
was found between hyperexpression of OCT4 (C5-fold,
relative to non-neoplastic control) and lower survival rate.
Median survival time of patients with tumors hyperexpressing OCT4 dropped to 6 months, while tumors displaying lower/undetectable OCT4 expression kept a
median survival of 153 months. Likewise, OCT4 tumor
expression levels in deceased patients were significantly
higher (P = 0.0169) than in patients still alive by the end
of the study (Fig. 3b). In an additional in silico analysis
with distinct microarray and SAGE datasets publically
available for medulloblastoma, OCT4 expression was also
found to be increased in tumors associated with poor
prognosis (supplementary material). No correlations were
found between detectable expression of either of these
three genes and metastasis. Since most medulloblastoma
cases were of the classic subtype, it was not possible to
evaluate associations between gene expression and histological subtypes.
OCT4 expression is correlated with poor prognosis
in clinically average risk patients
The clinicopathological data of patients with OCT4?
tumors is detailed in Table 2. Among them, nine (50%)
123
74
J Neurooncol (2012) 106:71–79
Table 1 Clinicopathological and gene expression features of medulloblastoma samples
Clinicopathological features
Number of cases
CD133 expression
LIN28 expression
OCT4 expression
Positive
Positive
Positive
Undetectable
Undetectable
Undetectable
Age at diagnosis (years)
\3
7 (18.9%)
6
1
6
1
3
4
30 (81.1%)
28
2
26
4
15
15
Male
26 (70.3%)
24
2
23
3
13
13
Female
11 (29.7%)
10
1
9
2
5
6
33 (89.2%)
30
3
28
5
16
17
Desmoplastic
1 (2.7%)
1
–
1
–
0
1
Large-cell
2 (5.4%)
2
–
2
–
1
1
Anaplastic
1 (2.7%)
1
–
1
–
1
–
0
–
–
–
–
–
–
Yes
13 (32.4%)
11
2
11
2
4
9
No
24 (67.6%)
23
1
21
3
16
8
Partial
13 (35.1%)
11
2
10
3
6
7
Total
23 (62.2%)
22
1
21
2
12
11
1 (2.7%)
1
–
1
–
–
1
[3 up to 18
Gender
Histological subtype
Classic
Extensive nodularity
Metastasis at diagnosis
Tumor resection
Not informative
Relapse
Yes
9 (24.3%)
7
2
7
2
5
4
No
Patient status
28 (74.7%)
27
1
25
3
13
15
Deceased
19 (51.3%)
17
2
18
1
12
7
Alive
18 (48.7%)
17
1
15
3
6
12
High
20 (54.1%)
18
2
17
3
9
11
Average
17 (45.9%)
16
1
15
2
9
8
Risk group
Fig. 1 Expression of stem cell-related genes in medulloblastoma.
Significant positive correlations were found between expression of
LIN28 and CD133 (a) as well as of OCT4 and LIN28 (b). The
respective Spearman’s Rank correlation coefficients and significance
123
levels are shown in the plots. Each dot corresponds to the concurrent
normalized gene expression level in each tumor sample, as determined by real-time PCR
J Neurooncol (2012) 106:71–79
75
Fig. 2 Expression of stem cell-related genes and survival of patients
with medulloblastoma. The Kaplan–Meier curves show overall
survival rates with respect to detectable OCT4, LIN28 or CD133
expression. Survival curves of patients classified based on clinical
prognostic criteria are also shown for comparison. Statistical differences were calculated by the log-rank test
were clinically classified as average risk, three of which
displaying OCT4 hyperexpressing tumors. When analyzing
overall survival within this clinical subgroup, there was
also a significant correlation between OCT4 expression and
lower survival rate (P = 0.0112; Fig. 3c). All average risk
patients deceased within 5 years of initial diagnosis had
OCT4? tumors, an outcome typical of most clinically high
risk patients. Median survival time of average risk patients
with OCT4? tumors was 28 months. Among these average
risk patients with shorter survival time, 60% had tumor
relapse.
From the stem cell markers evaluated in this study, only
OCT4 was found to be significantly correlated with prognostic factors. Not all tumor specimens had detectable
expression of CD133, LIN28 or OCT4. In fact, OCT4 was
the most rarely expressed gene, being detected in less than
a half the tumor collection examined here.
Information regarding expression of these stem cellrelated genes in medulloblastoma is rather limited. Highly
tumorigenic CD133? stem-like cells have originally been
isolated from medulloblastoma [5]. In the DAOY medulloblastoma cell line, CD133? cells were reported to be
more resistant to radiation than CD133-negative cells [11].
Furthermore, in oligodendroglial tumors and glioblastomas, CD133 expression is an indicator of shorter survival
[12, 13]. Based on what is known for these high-grade
gliomas, association between CD133 expression and poor
prognosis was also expected in medulloblastoma. In our
study, however, expression of this neural stem cell marker
was not found to be correlated with either metastasis or
survival.
In normal tissue, OCT4 and LIN28 expression are more
restricted to embryonic stem cells (ESC) and early embryonic structures, decreasing dramatically with the ESC differentiation [14, 15]. However, several recent studies have
Discussion
The clinicopathological characteristics of the patient cohort
included in the study were typical. Clinical stratification of
patients yielded 54% as high risk and 46% as average risk.
In agreement with the literature, the 5-year survival rates
for high and average risk patients were 33 and 70%,
respectively. Patient stratification solely based on clinical
parameters was not enough to prospectively identify
average risk patients with poor prognosis, once more
reinforcing the need for additional indicators of prognosis.
123
76
Fig. 3 Correlation of OCT4 expression levels and poor survival.
a Kaplan–Meier curve showing overall survival rates according to the
level of OCT4 expression (high or low/undetectable). b Median
normalized OCT4 expression levels with respect to patient status.
Only OCT4? tumors were included in this analysis. c Overall
survival rates according to OCT4 expression status (positive or
undetectable) in the subgroup of clinically average risk patients. All
patients with OCT4? tumors were deceased within 5 years of initial
diagnosis. In sharp contrast, all patients whose tumors had no
detectable expression of OCT4 were still alive during the same time
frame. Significant correlations with overall survival and with patient
status were determined by the log-rank and Mann–Whitney tests,
respectively
123
J Neurooncol (2012) 106:71–79
reported the involvement of both OCT4 and LIN28, as well
as other embryonic associated genes, with tumor formation
[16–19]. Although both genes are required for pluripotency
and, despite the fact that all OCT4? medulloblastoma
specimens had concomitant expression of LIN28, only the
former gene was significantly correlated with poor overall
survival. Of note, OCT4 expression was not restricted to a
particular clinical staging, being detected at a similar frequency in tumor specimens from either average or high risk
patients. In both clinical groups, patients whose tumors had
any detectable expression of OCT4 had a significant lower
survival time than patients with OCT4-negative tumors.
Moreover, the higher the relative OCT4 expression level,
the lower the survival time.
Of particular interest was the OCT4 expression pattern
in clinically average risk patients. Considering the 5-year
survival rate calculated for this clinical group in our cohort,
all deceased patients (corresponding to 30% of total average risk patients) had OCT4? tumors, while most of the
living patients had OCT4-negative tumors. The practical
meaning of this finding is that, in terms of 5-year survival
rating, average risk patients with OCT4? tumors had a
clinical outcome comparable to that of typical high risk
patients, displaying a median survival time of 28 months.
The positive correlation found between OCT4 expression and poor survival in medulloblastoma is in agreement
with recent studies in other cancers, where expression of
OCT4 and other pluripotency genes is correlated with
poorly differentiated and more aggressive tumors.
Although not previously addressed in medulloblastomas,
expression of OCT4 was reported to be associated with
higher grade gliomas, namely glioblastoma, when compared with oligodendroglioma and astrocytomas of grades
II and III [20]. Similarly, expression of OCT4 and NANOG,
another pluripotency-related gene, were detected in high
grade lung adenocarcinoma, and their ectopic expression in
tumor cells induced stem cell-like properties and enhanced
tumorigenicity [21]. Interestingly, an opposed association
between OCT4 levels and prognosis was found in hypopharyngeal squamous cell carcinoma, where it was correlated with lower tumor recurrence and longer patient
survival [22]. Thus, both expression and predictive value of
OCT4 may vary according to the type of cancer and the
expressing cell.
In human ESC, down-regulation of LIN28 expression
reduced OCT4 protein levels suggesting that LIN28 can
regulate OCT4 expression at a post-transcriptional level
[23]. In our study, although all OCT4? medulloblastoma
samples also expressed LIN28, several tumor specimens
had detectable expression of LIN28 but not of OCT4,
suggesting alternative mechanisms for OCT4 expression
regulation in this type of cancer. In murine neural stem
cells, a decreased expression of OCT4 along the embryonal
180
18
194
156
3
4
5
6
77
216
156
48
72
122
200
89
28
29
30
31
32
33
34
20
35
25
26
27
35
84
20
24
108
19
24
198
18
23
84
17
62
93
16
118
76
15
22
62
14
21
64
138
10
11
13
177
78
9
12
99
156
8
0.76
108
2
7
3
Age at
diagnosis
(months)
1
Patient
Large-cell
Classic
Classic
Classic
Desmoplastic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Large-cell
Anaplastic
Classic
Classic
Classic
Classic
Classic
Classic
Classic
Histological
subtype
68
75
97
103
123
0.5
41
63
4
15
6
19
41
41
153
31
59
27
45
61
40
94
10
60
18
24
11
5
28
6
3
10
2
7
Survival
(months)
M0
M0
M0
M0
M0
NI
NI
M0
NI
M0
M3
M3
M1
M2
M3
M1
M0
M0
M0
M0
M1
M0
M0
M0
M0
M0
M0
M0
M0
M0
M2
M0
M1
M1
Metastasis
T
T
T
T
T
T
NI
P
P
T
P
P
P
P
P
T
T
P
P
T
T
T
P
P
T
T
T
T
T
T
P
T
T
P
Tumor
resection
Table 2 Clinicopathological parameters of patients stratified by OCT4 expression
C?V?CP
C?V?CP
C?V?CP
IFO?ETO
IFO?ETO
POD
NI
C?V?CP?ETO
POD
C?V?CP?ETO
POD
POD
C?V?CP?ETO
C?V?CP?ETO
EP
NI
C?V?CP
NI
EP
C?V?CP
C?V?CP
C?V?CP
EP
EP
C?V?CP
C?V?CP
C?V?CP
POD
IFO?ETO
POD
POD
C?V?CP
POD
POD
Treatment
No
No
No
No
No
No
Yes
No
No
No
No
Yes
No
No
Yes
Yes
No
Yes
No
No
Yes
No
No
No
No
Yes
No
No
Yes
No
No
Yes
No
No
Tumor
recurrence
A
A
A
A
A
D
D
A
D
A
D
D
A
A
D
D
A
D
A
A
D
A
D
A
A
D
D
D
D
D
D
D
D
D
Status
AR
AR
AR
AR
AR
HR
HR
HR
HR
HR
HR
HR
HR
HR
HR
HR
AR
HR
HR
AR
HR
AR
HR
HR
AR
AR
AR
HR
AR
AR
HR
AR
HR
HR
Clinical
risk
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
7.78
39.90
76.36
93.10
155.41
311.79
504.17
1,064.09
1,018.46
1,120.18
1,354.15
3,861.32
5,263.13
7,372.39
9,662.98
21,756.45
22,715.48
317,464.43
OCT4
expression
(910-7)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
0.01
0.06
0.12
0.14
0.24
0.49
0.80
1.68
1.61
1.77
2.14
6.11a
8.33a
11.67a
15.30a
34.45a
35.97a
502.69a
Relative OCT4
expression
(OCT4/normal)
J Neurooncol (2012) 106:71–79
77
123
123
P partial, T total, D deceased, A alive, HR high risk, AR average risk, V vincristine, C cisplatin, CP cyclophosphamide, ETO etoposide, IFO ifosfamide, EP experimental protocol, POD
postoperative death, ND not detectable, NI not informed
a
OCT4 hyperexpressing tumors
–
–
ND
AR
No
91
37
Classic
41
M0
T
C?V?CP
A
ND
AR
ND
AR
No
A
No
T
C?V?CP
M0
M0
52
50
98
36
Classic
112
35
Classic
T
C?V?CP
A
OCT4
expression
(910-7)
Clinical
risk
Status
Tumor
recurrence
Treatment
Tumor
resection
Metastasis
Survival
(months)
Histological
subtype
Age at
diagnosis
(months)
Patient
Table 2 continued
–
J Neurooncol (2012) 106:71–79
Relative OCT4
expression
(OCT4/normal)
78
development has been associated with a gradual methylation in its promoter region [24]. Further studies on the
methylation status of OCT4 promoter should help understand the expression pattern found for medulloblastoma
patients with poor prognosis.
In conclusion, our findings reveal a predictive prognostic value for OCT4 expression in medulloblastoma.
Expression of OCT4, but not CD133 or LIN28 stem cellrelated genes, was significantly correlated with shorter
survival in either clinically high or average risk patients.
More importantly, positive OCT4 expression in primary
tumor specimens could discriminate patients clinically
classified as average risk but which deceased within
5 years of diagnosis, a clinical outcome typical of high risk
patients. Thus, OCT4 expression status might be used as a
further indicator of poor prognosis to aid postoperative
treatment selection, with a particular potential benefit for
clinically average risk patients.
Acknowledgments This work was supported by grants from INCTCélulas Tronco em Doenças Genéticas Humanas, FAPESP, CNPq,
and CAPES. C.O.R. and D.E.S. were recipients of fellowships from
CAPES and CNPq.
References
1. Guessous F, Li Y, Abounader R (2008) Signaling pathways in
medulloblastoma. J Cell Physiol 217:577–583
2. Kouprina N, Pavlicek A, Collins NK et al (2005) The microcephaly
ASPM gene is expressed in proliferating tissues and encodes for a
mitotic spindle protein. Hum Mol Genet 14:2155–2165
3. Mueller S, Chang S (2009) Pediatric brain tumors: current
treatment strategies and future therapeutic approaches. Neurotherapeutics 6:570–586
4. Carlotti CG Jr, Smith C, Rutka JT (2008) The molecular genetics
of medulloblastoma: an assessment of new therapeutic targets.
Neurosurg Rev 31:359–369
5. Singh SK, Clarke ID, Terasaki M et al (2003) Identification of a
cancer stem cell in human brain tumors. Cancer Res 63:5821–5828
6. Okamoto OK (2009) Cancer stem cell genomics: the quest for
early markers of malignant progression. Expert Rev Mol Diagn
9:545–554
7. Cheng L, Sung MT, Cossu-Rocca P et al (2007) OCT4: biological functions and clinical applications as a marker of germ cell
neoplasia. J Pathol 211:1–9
8. Gidekel S, Pizov G, Bergman Y, Pikarsky E (2003) Oct-3/4 is a
dosedependent oncogenic fate determinant. Cancer Cell 4:
361–370
9. Chang CH, Housepian EM, Herbert C (1969) An operative
staging system and a megavoltage radiotherapeutic technic for
cerebellar medulloblastoma. Radiology 93:1351–1359
10. Valente V, Teixeira SA, Neder L et al (2009) Selection of suitable housekeeping genes for expression analysis in glioblastoma
using quantitative RT-PCR. BMC Mol Biol 3:10–17
11. Blazek ER, Foutch JL, Maki G (2007) Daoy medulloblastoma
cells that express CD133 are radioresistant relative to CD133cells, and the CD133? sector is enlarged by hypoxia. Int J Radiat
Oncol Biol Phys 67:1–5
J Neurooncol (2012) 106:71–79
12. Beier D, Wischhusen J, Dietmaier W et al (2008) CD133
expression and cancer stem cells predict prognosis in high-grade
oligodendroglial tumors. Brain Pathol 18:370–377
13. Zhang M, Song T, Yang L et al (2008) Nestin and CD133:
valuable stem cell-specific markers for determining clinical outcome of glioma patients. J Exp Clin Cancer Res 24:27–85
14. Richard M, Tan SP, Tan JH et al (2004) The transcriptome profile
of human embryonic stem cells as defined by SAGE. Stem Cells
22:51–64
15. Darr H, Benvenisty N (2009) Genetic analysis of the role of the
reprogramming gene LIN-28 in human embryonic stem cells.
Stem Cells 27:352–362
16. Marie SK, Okamoto OK, Uno M et al (2008) Maternal embryonic
leucine zipper kinase transcript abundance correlates with malignancy grade in human astrocytomas. Int J Cancer 122:807–815
17. Viswanathan SR, Powers JT, Einhorn W et al (2009) Lin28
promotes transformation and is associated with advanced human
malignancies. Nat Genet 41:843–848
18. Lu L, Katsaros D, Shaverdashvili K et al (2009) Pluripotent factor
lin-28 and its homologue lin-28b in epithelial ovarian cancer and
their associations with disease outcomes and expression of let-7a
and IGF-II. Eur J Cancer 45:2212–2218
79
19. Dangi-Garimella S, Yun J, Eves EM et al (2009) Raf kinase
inhibitory protein suppresses a metastasis signalling cascade
involving LIN28 and let-7. EMBO J 28:347–358
20. Ben-Porath I, Thomson MW, Carey VJ et al (2008) An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet 40:499–507
21. Chiou SH, Wang ML, Chou YT et al (2010) Co-expression of
Oct4 and Nanog enhances malignancy in lung adenocarcinoma
by inducing cancer stem cell-like properties and epithelial-mesenchymal transdifferentiation. Cancer Res 70:10433–10444
22. Ge N, Lin HX, Xiao XS et al (2010) Prognostic significance of
Oct4 and Sox2 expression in hypopharyngeal squamous cell
carcinoma. J Transl Med 8:94
23. Qiu C, Ma Y, Wang J et al (2010) Lin28-mediated post-transcriptional regulation of Oct4 expression in human embryonic
stem cells. Nucleic Acids Res 38:1240–1248
24. Lee SH, Jeyapalan JN, Appleby V et al (2010) Dynamic methylation and expression of Oct4 in early neural stem cells. J Anat
217:203–213
123