Sage Journals: Discover world-class research

Abstract

BACKGROUND:

The treatment strategies of astrocytoma have not changed considerably due to the restricted appreciation of its biology.

OBJECTIVES:

This study aimed to evaluate the expression of the stem cell-related proteins (CD133 and Sox2) and their prognostic value in astrocytic glioma.

METHODS:

The immunohistochemical expression of CD133 and Sox2 in 40 patients with an astrocytic glioma of different grades was studied. The recorded data on the overall survival (OS), progression-free survival (PFS) and the response to the therapeutic protocol were collected and lastly analyzed.

RESULTS:

CD133 expression was observed in 87.5% of the cases, while positive Sox2 expression was found in all the studied cases. There was a significant association of CD133 expression with the histological grade and the tumor size ( $p<$ 0.001). A significant association of Sox2 with the histological grade and the tumor size was noted ( $p=$ 0.004, $p=$ 0.006 respectively). Up-regulation of both CD133 and Sox2 had a significant association with poor clinical response to the therapy ( $p<$ 0.001 for each). Shorter OS and PFS were related to CD133 and Sox2 overexpression.

CONCLUSIONS

: Astrocytoma with CD133 and Sox2 overexpression had an unfavorable prognosis and poor clinical response to the current therapeutic protocol.

Keywords

CD133 Sox2 astrocytoma immunohistochemistry

1. Introduction

Astrocytoma is the most prevalent intracranial tumor that accounts for about 75% of neuroepithelial tumors. They are classified according to the WHO scale into low-grade astrocytoma (GI-II), anaplastic astrocytoma (GIII), and glioblastoma multiforme (GIV) [1]. In Egypt, astrocytic glioma is the most common glial tumor (79.4% of all gliomas) [2].

Regardless of the recent developments, the majority of astrocytic glioma in remains resistant to therapy. Recent studies suggested that the tumor biology and the resistance to treatment are attributed to the presence of cancer stem cells (CSCs). CSCs are defined as a subpopulation of cancer cells that have the capacity for indefinite self-renewal and the generation of different types of progeny through differentiation [3].

The importance of CSCs in estimating the prognosis of patients with gliomas has been evaluated using several markers closely related to their presence [4], and the search for novel prognostic or predictive molecular parameters are still the primary target for the improvement of its clinical management [5].

CD133 is the marker of choice for identifying cancer stem cells in gliomas [6]. It plays a fundamental role in cancer progression by the maintenance of stem cell properties of cancer cells, such as their ability to self-renew. It has been supported that the expression of CD133 is controlled by epigenetic factors [7].

Sox2 is a transcription factor belonging to the sex-determining region Y-box family, which is expressed in a wide array of tissues and play important roles in the regulation of organ development, cell type specificity, and in the pluripotency maintenance of CSCs of self-renewal and differentiation [8, 9].

Sox2 overexpression has been reported in a growing list of tumors, including esophageal carcinoma, pancreatic carcinoma, lung cancer, breast cancer, ovarian carcinoma, hepatocellular carcinoma and head and neck cancers [1]. Particularly, the Sox2 expression is important for the maintenance and development of the CNS tumors and positively associated with the tumor grade of brain tumors [10, 11].

Therefore, the idea of the present study was to evaluate the expression level of these CSC-related proteins in different astrocytic glioma grades and to assess their predictive role regarding the survival and response to therapy.

2. Materials and methods

2.1 Patients and tumor specimens

We searched the database of the Pathology department and identified 40 Formalin-fixed, paraffin-embedded (FFPE) tissue specimens of primary intracranial astrocytoma. Surgical resection was done in the Neurosurgery department of the same institution, during the period from January 2013 to December 2015. The clinical data of the studied cases were collected from the patients’ medical files, including the age, sex, preoperative Karnofsky performance status (KPS) score, treatment strategies, follow-up data, postoperative survival data, and records of the response to therapy. We exclude patients that had a recurrent tumor, incomplete clinical information or uncertain follow-up data. The hematoxylin and eosin stained slides were reviewed and the cases were graded in accordance with the 2016 World Health Organization classification of tumors of the central nervous system [12]. The study was approved by the ethics committee of Zagazig University Hospitals.

According to the treatment protocol, patients with grade II astrocytoma get a three-dimensional conformal radiotherapy of 45–54 Gy in 1.8–2 Gy fractions. Grade-III astrocytoma received a three-dimensional conformal radiotherapy of 60 Gy in two Gy fractions, while GIV astrocytoma (GBM) received a chemo-radiotherapy regimen including temozolomide 75 mg/m ${}^{2}$ daily with a three-dimensional conformal radiotherapy of 60 Gy in two Gy fractions, followed by 6 cycles of temozolomide 150 mg/m ${}^{2}$ for five days each 28 days.

Follow-up of the patients was done by magnetic resonance image (MRI) every 3–4 months to detect the patients’ response. A complete response (CR) was defined as a complete disappearance of the tumor; partial response (PR) was $>$ 50% shrinkage of the tumor area; stable disease (SD) was measured as $\leqslant$ 50% shrinkage of the tumor area. An increase of $>$ 25% of the tumor area was considered as progressive disease (PD).

2.2 Immunohistochemistry

Immunohistochemical staining was carried out on a Dako autostainer universal staining system (Dako Denmark A/S, Glostrup, Denmark) by the avidin-biotin method. One paraffin-embedded block of astrocytoma tissue was selected from each case and cut into 4- $\mu$ m thick sections. The sections were subjected to dewaxing, rehydration, blocking with hydrogen peroxide, and antigen retrieval with the microwave. Then, the slides were incubated overnight at 2–8 ${}^{\circ}$ C with primary antibodies: anti-CD133 (rabbit polyclonal antibody Clone: MSTP061, Cat # 137634; Lab Vision); anti-Sox2 (1:400; rabbit polyclonal antibody, MA, USA). After washing in PBS, a biotin-marked secondary antibody was applied for 10 min then a peroxidase-marked streptavidin for an additional 10 min. The reaction was visualized by diaminobenzidine. Counterstaining with Mayer’s hematoxylin was done and then washed with distilled water. Negative controls, obtained by replacement of primary antibodies by PBS, were included within each slide batch.

2.3 Assessment of the immunohistochemistry

Immunohistochemical staining for CD133 and Sox2 was scored for the tumor cells. The number of positively stained cells in ten representative microscopic fields was counted and the percentage of positive cells was calculated. The results were interpreted by two independent pathologists who were blind to the clinicopathological and follow up data.

Figure 1.

A) Diffuse astrocytoma (GII) (H&E, $\times$ 100), B) Gemistocytic astrocytoma (GII) (H&E, $\times$ 400), C) Anaplastic astrocytoma (GIII) (H&E, $\times$ 200), D) Glioblastoma (GIV) (H&E, $\times$ 200).

2.3.1 CD133 scoring

The cytoplasmic/or membranous staining of CD133 was considered as a positive expression. CD133 staining was semi-quantitatively graded for the percentage of tumor cells: negative $<$ 10% of the tumor cells were stained, focal: 10–50% of the tumor cells was stained, and diffuse staining if $\geqslant$ 50% of the tumor cells were stained [13].

2.3.2 Sox2 scoring

We classified the percentage of Sox2 expression in tumor cells as low ( $<$ 20% of cells expressing Sox2), moderate (20%–60% of tumor cells expressing Sox2), and high ( $>$ 60% of tumor cells expressing Sox2) [14].

2.4 Statistics

Continuous variables were expressed as the mean $\pm$ SD and median (range), and the categorical variables were expressed as a number (percentage). Continuous variables were checked for normality by using Shapiro-Wilk test. Kruskal Wallis H test was used to compare between more than two groups of normally distributed variables. Percent of categorical variables were compared using the Pearson’s Chi-square test. A trend of change in the distribution of relative frequencies between ordinal data was compared using Chi-square test for trend. Progression Free Survival (PFS) was calculated as the time from start of radiotherapy to the time of progression or the most recent follow-up in which patient was progression-free (censored). Overall Survival (OS) was calculated as the time from diagnosis to death or the most recent follow-up contact (censored). Stratification of PFS and OS was done according to immunohistochemical staining for SOX2 and CD133. These time-to-event distributions were estimated using the method of Kaplan-Meier plot and compared using two-sided exact log-rank test. All tests were two-sided. A $p$ -value $<$ 0.05 was considered significant. All statistics were performed using SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA) and MedCalc windows (MedCalc Software bvba 13, Ostend, Belgium).

Table 1
Clinicopathological features of 40 astrocytic glioma patients

Characteristics	All ( $N=$ 40)
	No.	%
Age (years)
Mean $\pm$ SD	46.47 $\pm$ 10.97
Median (range)	50 (20–64)
$\leqslant$ 50 years	23	57.5%
$>$ 50 years	17	42.5%
Sex
Male	27	67.5%
Female	13	32.5%
Surgery
GTR	24	60%
STR	10	25%
Biopsy	6	15%
KPS
$>$ 50–70%	25	62.5%
$>$ 70%	15	37.5%
Tumor size (cm)
Mean $\pm$ SD	4.59 $\pm$ 1.28
Median (range)	4.70 (2–9.40)
$\leqslant$ 3 cm	4	10%
$>$ 3 cm	36	90%
Grade
Grade II	12	30%
Grade III	10	25%
Grade IV	18	45%
Residual disease
Absent	0	0%
Present	40	100%
Sox2 intensity
Low	11	27.5%
Moderate	12	30%
High	17	42.5%
CD133 extension
Negative	5	12.5%
Focal positive	11	27.5%
Diffuse positive	24	60%
Radiotherapy
No	0	0%
Yes	40	100%
Radiotherapy dose
50–50.4 Gy	6	15%
54 Gy	6	15%
59.4–60 Gy	28	70%
Chemotherapy
No	22	55%
Yes	18	45%
Response
CR	3	7.5%
PR	13	32.5%
SD	11	27.5%
PD	13	32.5%
OAR	16	40%
NR	24	60%
Follow-up (months)
Mean $\pm$ SD	24.65 $\pm$ 13.31
Median (range)	21.50 (8–48)
Progression	( $N=$ 24)
Absent	14	58.3%
Present	10	41.7%

3. Results

3.1 Patients’ characteristics

The present study included forty patients; 27 males and 13 females. The mean age of the patients at the initial diagnosis was 46.47 $\pm$ 10.97 (range 20–64 years). The studied cases were classified into astrocytoma GII (12 cases (30%); 5 cases gemistocytic astrocytoma and 7 cases diffuse astrocytoma), anaplastic astrocytoma GIII (10 cases, 25%), and 18 cases of glioblastoma (GIV, 45%) (Fig. 1). The majority of the tumor masses (60%) was operated by gross total resection (GTR), where most of the specimens (90%) were $>$ 3 cm.

After the treatment, according to the protocol an overall response (OAR) of 40% of the studied cases was noted, while 60% exhibited no clinical response (NR). The median follow-up duration was 21.5 months (range 8–48 months), during which 17 patients (42.5%) died and 10 cases out of 24 cases (41.7%) that revealed no response had a progression. The clinicopathological features and the treatment regimen of our cases were presented in Table 1.

Table 1, continued
Characteristics	All ( $N=$ 40)
	No.	%
Mortality
Absent	17	42.5%
Present	23	57.5%

Continuous variables were expressed as mean $\pm$ SD and median (range); categorical variables were expressed as number (percentage).

3.2 The association of CD133 and Sox2 expression with the clinicopathological features

The cytoplasmic/or membranous expression of CD133 was detected in 87.5% of the studied cases, where 60% revealed diffuse immunostaining. There was a significant up-regulation of CD133 expression with the higher astrocytoma grade ( $p<$ 0.001) and larger tumor size ( $p=$ 0.014). There was a non-significant association of CD133 expression with the patient age ( $p=$ 0.135) or patient gender ( $p=$ 0.872) (Table 2, Fig. 2).

The nuclear expression of Sox2 was observed in all the studied cases, where 42.5% revealed high expression. There was a significant up-regulation of Sox2 expression with the higher astrocytoma grade ( $p=$ 0.004) and larger tumor size ( $p=$ 0.006). There was non-significant association of Sox2 expression with the patient age ( $p=$ 0.171) or patient gender ( $p=$ 0.446) (Table 3, Fig. 3).

There was a significant association between SOX2 and CD133 expression ( $p<$ 0.001); 63.6% of low SOX2 cases had a focal CD133 expression and 94.1% of high SOX2 expression had a diffuse CD133 expression.

Table 2
Relations between clinicopathological features and CD133 expression

	All ( $N=$ 40)		Negative ( $N=$ 5)		Focal positive ( $N=$ 11)		Diffuse positive ( $N=$ 24)		$p$ -value
			CD133 expression
Characteristics	No.	%	No.	%	No.	%	No.	%
Age (years)
Mean $\pm$ SD	46.47 $\pm$ 10.97		44.40 $\pm$ 13.61		38.72 $\pm$ 12.52		50.45 $\pm$ 7.58		0.033 $\bullet$
Median (range)	50 (20–64)		38 (29–62)		37 (20–57)		51.50 (30–64)
$\leqslant$ 50 years	23	57.5%	3	13%	9	39.1%	11	47.8%	0.135‡
$>$ 50 years	17	42.5%	2	11.8%	2	11.8%	13	76.5%
Sex
Male	27	67.5%	3	11.1%	8	29.6%	16	59.3%	0.872‡
Female	13	32.5%	2	15.4%	3	23.1%	8	61.5%
Tumor size (cm)
Mean $\pm$ SD	4.59 $\pm$ 1.28		3.66 $\pm$ 1.41		3.72 $\pm$ 0.82		5.19 $\pm$ 1.11		0.001 $\bullet$
Median (range)	4.70 (2–9.40)		3.50 (2–5.50)		3.90 (2.50–5.30)		5 (3.80–9.40)
$\leqslant$ 3 cm	4	10%	2	50%	2	50%	0	0%	0.014‡
$>$ 3 cm	36	90%	3	8.3%	9	25%	24	66.7%
Grade
Grade II	12	30%	4	33.3%	6	50%	2	16.7%	$<$ 0.001§
Grade III	10	25%	1	10%	3	30%	6	60%
Grade IV	18	45%	0	0%	2	11.1%	16	88.9%
Sox2 expression
Low	11	27.5%	3	27.3%	7	63.6%	1	9.1%	$<$ 0.001§
Moderate	12	30%	2	16.7%	3	25%	7	58.3%
High	17	42.5%	0	0%	1	5.9%	16	94.1%

Categorical variables were expressed as number (percentage); continuous variables were expressed as mean $\pm$ SD and median (range); $\bullet$ Kruskal Wallis H test; ‡ Chi-square test; § Chi-square test for trend; $p<$ 0.05 is significant.

Figure 2.

CD133 cytoplasmic immunoexpression in astrocytic glioma cases; A) Focal CD133 immunoexpression in a case of diffuse astrocytoma (IHC, $\times$ 100), B) Diffuse CD133 immunoexpression in a case of gemistocytic astrocytoma (IHC, $\times$ 400), C) Diffuse CD133 immunoexpression in a case of anaplastic astrocytoma (IHC, $\times$ 400), D) Diffuse CD133 immunoexpression in a case of glioblastoma (IHC, $\times$ 100).

Table 3

Relations between clinicopathological features and Sox2 expression

	All ( $N=$ 40)		Low ( $N=$ 11)		Moderate ( $N=$ 12)		High ( $N=$ 17)		$p$ -value
			SOX2 expression
Characteristics	No.	%	No.	%	No.	%	No.	%
Age (years)
Mean $\pm$ SD	46.47 $\pm$ 10.97		43.09 $\pm$ 13.54		42.50 $\pm$ 11.01		51.47 $\pm$ 7.01		0.067 $\bullet$
Median (range)	50 (20–64)		38 (25–62)		43 (20–60)		52 (30–64)
$\leqslant$ 50 years	23	57.5%	7	30.4%	9	39.1%	7	30.4%	0.171‡
$>$ 50 years	17	42.5%	4	23.5%	3	17.6%	10	58.8%
Sex
Male	27	67.5%	9	33.3%	8	29.6%	10	37%	0.446‡
Female	13	32.5%	2	15.4%	4	30.8%	7	58.3%
Tumor size (cm)
Mean $\pm$ SD	4.59 $\pm$ 1.28		3.66 $\pm$ 1.16		4.76 $\pm$ 1.62		5.08 $\pm$ 0.70		0.006 $\bullet$
Median (range)	4.70 (2–9.40)		3.50 (2–5.50)		4.45 (4–9.40)		5 (3.80–6.10)
$\leqslant$ 3 cm	4	10%	3	75%	1	25%	0	0%	0.062‡
$>$ 3 cm	36	90%	8	22.2%	11	30.6%	17	47.2%
Grade
Grade II	12	30%	7	58.3%	3	25%	2	16.7%	0.004§
Grade III	10	25%	2	20%	4	40%	4	40%
Grade IV	18	45%	2	11.1%	5	27.8%	11	61.1%
CD133 expression
Negative	5	12.5%	3	60%	2	40%	0	0%	$<$ 0.001§
Focal positive	11	27.5%	7	63.6%	3	27.3%	1	9.1%
Diffuse positive	24	60%	1	4.2%	7	29.2%	16	66.7%

Figure 3.

Sox2 nuclear immunoexpression in astrocytic glioma cases; A) Low Sox2 immunoexpression in a case of diffuse astrocytoma (IHC, $\times$ 400), B) High Sox2 immunoexpression in a case of anaplastic astrocytoma (IHC, $\times$ 400), C) High Sox2 immunoexpression in a case of glioblastoma (IHC, $\times$ 400).

Table 4

Relations between CD133 expression and treatment outcome

	All		Negative		Focal positive		Diffuse positive		$p$ -value
			CD133 expression
Characteristics	No.	%	No.	%	No.	%	No.	%
Response	( $N=$ 40)		( $N=$ 5)		( $N=$ 11)		( $N=$ 24)
CR	3	7.5%	2	40%	1	9.1%	0	0%	$<$ 0.001§
PR	13	32.5%	3	60%	9	81.8%	1	4.2%
SD	11	27.5%	0	0%	1	9.1%	10	41.7%
PD	13	32.5%	0	0%	0	0%	13	54.2%
OAR	16	40%	5	100%	10	90.9%	1	4.2%	$<$ 0.001§
NR	24	60%	0	0%	1	9.1%	23	95.8%
Progression	( $N=$ 24)		( $N=$ 3)		( $N=$ 10)		( $N=$ 11)
Absent	14	58.3%	3	100%	9	90%	2	18.2%	$<$ 0.001§
Present	10	41.7%	0	0%	1	10%	9	81.8%
PFS
Median (months)	NR		NR		NR		10 months		$<$ 0.001†
1-year PFS	74.4%		100%		100%		45.5%
2-year PFS	60.1%		100%		100%		11.4%
3-year PFS	55.1%		100%		87.5%		–
Mortality	( $N=$ 40)		( $N=$ 5)		( $N=$ 11)		( $N=$ 24)
Absent	17	42.5%	5	100%	10	90.9%	2	8.3%	$<$ 0.001§
Present	23	57.5%	0	0%	1	9.1%	22	91.7%
OS
Median (months)	24 months		NR		NR		14 months		$<$ 0.001†
1-year OS	79.8%		100%		100%		58.3%
2-year OS	50.2%		100%		100%		9.1%
3-year OS	38.2%		100%		87.5%		–

Categorical variables were expressed as number (percentage); NR: Not reach yet; § Chi-square test for trend; † Log-rank test; $p<$ 0.05 is significant.

Table 5

Relations between Sox2 expression and treatment outcome

	All		Low		Moderate		High		$p$ -value
			Sox2 expression
Characteristics	No.	%	No.	%	No.	%	No.	%
Response	( $N=$ 40)		( $N=$ 11)		( $N=$ 12)		( $N=$ 17)
CR	3	7.5%	3	27.3%	0	0%	0	0%	$<$ 0.001§
PR	13	32.5%	8	72.7%	5	41.7%	0	0%
SD	11	27.5%	0	0%	7	58.3%	4	23.5%
PD	13	32.5%	0	0%	0	0%	13	76.5%
OAR	16	40%	11	100%	5	41.7%	0	0%	$<$ 0.001§
NR	24	60%	0	0%	7	58.3%	17	100%
Progression	( $N=$ 24)		( $N=$ 8)		( $N=$ 12)		( $N=$ 4)
Absent	14	58.3%	8	100%	5	41.7%	1	25%	0.006§
Present	10	41.7%	0	0%	7	58.3%	3	75%
PFS
Median (months)	NR		NR		24 months		15 months		0.019†
1-year PFS	74.4%		100%		58.3%		66.7%
2-year PFS	60.1%		100%		50%		0%
3-year PFS	55.1%		100%		41.7%		0%
Mortality	( $N=$ 40)		( $N=$ 11)		( $N=$ 12)		( $N=$ 17)
Absent	17	42.5%	11	100%	5	41.7%	1	5.9%	$<$ 0.001§
Present	23	57.5%	0	0%	7	58.3%	16	94.1%
OS
Median (months)	24 months		NR		24 months		17 months		$<$ 0.001†
1-year OS	79.8%		100%		75%		69.7%
2-year OS	50.2%		100%		50%		19%
3-year OS	38.2%		100%		37.5%		0%

Categorical variables were expressed as number (percentage); NR: Not reach yet; § Chi-square test for trend; † Log-rank test; $p<$ 0.05 is significant.

Figure 4.

Kaplan-Meier plot for Progression Free Survival (PFS); A) Stratified according to CD133 IHC staining ( $p<$ 0.001) and B) Stratified according to Sox2 IHC staining ( $p=$ 0.019).

Figure 5.

Kaplan-Meier plot for Overall Survival (OS): A) Stratified according to CD133 IHC staining ( $p<$ 0.001), and B) Stratified according to Sox2 IHC staining ( $p<$ 0.001).

3.3 Prognostic relevance of CD133 and Sox2 expression in astrocytic glioma (Tables 4 and 5)

Regarding the response to the therapy, only 3 cases (7.5%) achieved a complete clinical response (CR), 13 cases (32.5%) had a partial response (PR) while the remaining cases exhibited no response (NR). CD133 up-regulation showed a significant association with the poor response to therapy ( $p<$ 0.001) and the tumor progression ( $p<$ 0.001). Furthermore, Sox2 overexpression showed a significant association with the poor response to therapy ( $p<$ 0.001) and the tumor progression ( $p=$ 0.006). At the end of the study follow-up period, 10/24 patients revealed a tumor progression after their failure to respond to the therapy protocol.

CD133 overexpression was significantly associated with shorter OS ( $p<$ 0.001) and PFS ( $p<$ 0.001). Likewise, Sox2 overexpression was significantly associated with shorter OS ( $p<$ 0.001) and PFS ( $p=$ 0.019). Kaplan-Meier plot curves that were stratified according to CD133 and SOX2 expression were presented in Fig. 5.

4. Discussion

Glioma is the most common type of primary brain tumor. Among them, anaplastic astrocytoma and glioblastoma are highly malignant with unfavorable prognosis. Only 10% of high-grade tumors survive for more than five years despite the current advances in the research and therapy [15, 16]. Therefore, improved understanding of the tumor biology, the molecular basis of invasion and spread, and resistance to standard therapeutic modalities may aid to improve patients’ outcome and the prediction of the future recurrence [17].

The identification of stem cells in brain tumors provides a powerful tool to investigate the tumorigenic process in the central nervous system and to develop targeted therapies to these cells. However, the methods for the isolation/enrichment of CSCs remain still imperfect and require improvements, as a consequence of the lack of universal markers [18].

In the last years, the treatment approaches for astrocytic glioma have not changed markedly because of the limited understanding of the biology of the tumor. The poor prognosis of the patients is attributed to the presence of drug-resistant cells which are insensitive to the current strategies [19]. Cancer stem cells (CSCs) comprise a subpopulation of tumor cells with a self-renewal capacity and ability for tumor initiation. CSCs are also responsible for the tumor recurrence and resistance to the current therapeutic modalities. Therefore, the identification of these cells seems to be fundamental for understanding the process of carcinogenesis, and for the development of targeted therapies [20].

In our study, CD133 expression was noted in 87.5% of the studied cases. Previously [21, 22], CD133 expression was reported in 77.6 and 96.2%, respectively of their astrocytoma cases. These different results may be attributed to the different specificity and sensitivity of the commercially available antibodies used in these studies. It was reported that different CD133 antibody clones identify different epitopes [11]. In addition, we observed a significant up-regulation of CD133 expression with astrocytoma grade and the tumor size as reported before [23]. Therefore, we supposed that CD133 overexpression is a marker of aggressive astrocytic glioma and a crucial factor for tumor progression.

In the current work, a significant association was present between CD133 overexpression and poor OS and PFS, where the patient survival declines significantly in those with diffuse CD133 expression. These results go with a previous investigation [23], which reported that the patients whose tumors expressed a high-level of CD133 had a mean survival time of 25.7 months compared to 45.7 months for patients whose tumors expressed a low-level of CD133. Similarly, another study reported a significant association with poor overall survival (OS) ( $p=$ 0.010) and PFS ( $p=$ 0.017) [13]. Therefore, we concluded that CD133 overexpression is an unfavorable prognostic marker in patients with astrocytic glioma. However, there is a considerable controversy about its prognostic significance in astrocytoma, where some studies [22, 24] revealed a non-significant association between CD133 expression and the patient survival. This controversy can be attributed to the different grades of astrocytoma included in these studies or the different scoring methods. The prognostic value of CD133 expression has been also determined in esophageal carcinoma [25], thyroid carcinoma [26], pancreatic carcinoma [27], and ovarian carcinoma [28].

Angelastro and Lamé supported the hypothesis that CD133 plays an antiapoptotic role in protecting CSCs from chemotherapeutic effect [29]. Moreover, CD133-positive glioma stem-like cells have the ability to survive high-dose irradiation leading to recurrent glioma [30]. Therefore, targeting CSCs could eradicate CNS tumors and protect against recurrent tumors. SirT1, which plays an important role in governing the radiosensitivity of glioma cells, is a downstream target of CD133 [31].

Our work reported a poor response to the therapy protocol in the cases that revealed diffuse CD133 expression, and this result matched with the previous reports [32, 33], which found that CD133 $+$ GBM and CSC CD133 $+$ demonstrate a high ability for radiation and chemotherapy resistance. Therefore, utilizing CD133 targeting in cancer therapy may open a window for new therapeutic opportunities. However, different studies [34, 35] found no relation between CD133 and the therapy resistance, but instead they claimed the resistance or the sensitivity to therapy to be a patient-specific. This controversy may be attributed to diverse scoring methods or heterogeneous expression of the stem cell markers.

Another marker evaluated in the present study was Sox2, an established stem cell regulator highly expressed in multiple tissue stem cells, including various types of neural stem cells and progenitor cells [36]. In addition, Sox2 has been demonstrated to be a glioma-specific marker [4].

Detectable Sox2 expression was found in all cases of the studied astrocytoma with different intensity grades, suggesting a role for increased Sox2 expression in glioma development, which was in agreement with Annovazzi et al. study [37]. Schmitz et al. found that Sox2 is overexpressed in all the investigated malignant glioma ranging from 6 to 66% stained tumor cells [38].

Our results revealed a positive relation between Sox2 expression and malignancy grade in astrocytoma which was consistent with other investigators [14, 39]. Similarly, Annovazzi et al. reported that the percentage of intensely Sox2 stained nuclei in glioblastoma were higher than diffuse astrocytoma [37]. These findings suggest the substantial contribution of Sox2 expression in the tumor progression.

Knockdown experiments have indicated that Sox2 is required to sustain the aggressive growth and infiltrative behavior of GBMs [40, 41]. In particular, Gangemi et al. reported that Sox2-silenced glioblastoma tumor-initiating cells have been shown to inhibit proliferation and tumorigenicity in immune-deficient mice, indicating that Sox2 may be a target for glioblastoma therapy [40]. Furthermore, Alonso et al. showed that the knockdown of the Sox2 gene in the GBM cell line reduces cellular proliferation and colony formation [42]. Several strategies are starting to use Sox2 directly or indirectly to target GSCs [43]. Therefore, these data suggest that Sox2 is a key gene that maintains the stemness of glioma stem cells [44].

In contrast, Phi et al. concluded that the expression of Sox2 did not correlate with malignancy grade in astrocytic tumors [45]. This discrepancy may due to heterogeneous nature the astrocytic tumors. Brain tumors are heterogeneous with respect to genetic and histological properties of cells within the tumor tissue [46]. Sox2 is known to be involved in the development and progression of multiple types of tumor [47, 48, 49].

High levels of Sox2 have been associated with tumor aggressiveness and worse prognosis [50]. Wang et al. [39], verified that patients with positive Sox2 expression exhibited a significantly reduced survival rate, compared with those with negative Sox2 expression ( $p<$ 0.001) which was parallel to our results as we found a significant relation between Sox2 and both OS and PFS. Thus, Sox2 might be of clinical value for risk stratification of astrocytoma. On the other hand, Wan et al. [52] and Galatoro et al. [51] reported that no impact of Sox2 on survival. This controversy may be due to the different number of examined patients; Wan et al. examined 283 astrocytomas WHO grade II–IV [52].

Regarding the response to therapy, our finding was similar to the preceding studies [43], where the cases with elevated Sox2 expression were more resistant to the therapy. However, Lemke et al. reported that the Sox2 positive fraction of glioma-initiating cell cultures was sensitive to irradiation [53]. The resistance to chemotherapy might be due to the prominent role of Sox2 in the regulation of GSCs [43]. Furthermore, Sox2 overexpression in cancer cells promoted evasion of apoptosis in tumor cells through antiapoptotic signaling [54]. Therefore, targeting its activity may offer an attractive therapeutic approach to treat glioblastoma.

The prior studies in ovarian cancer cell lines and breast cancer cell lines [55, 56], have shown that exogenous elevation of Sox2 can promote resistance to chemotherapeutics currently used. In addition, silencing Sox2 efficiently suppressed the expression of drug-resistance and anti-apoptotic genes and increased the sensitivity of the cells to radiation in oral squamous cell carcinoma [57].

Immunotherapy represents a promising treatment alternative to improve the clinical outcome of patients suffering from glioblastoma [43]. Schmitz et al. recognized Sox2 as a glioma-associated antigen abundantly and specifically overexpressed in glioma cells. In addition, they identified an immunogenic HLA-A*0201-restricted T-cell epitope derived from Sox2 that effectively activated tumor-directed cytotoxic T lymphocytes. Therefore, they suggested Sox2 as a promising novel strategy based on immunotherapy [38].

Interestingly, Song et al. found that Sox2 regulates tumor-initiating and drug-resistant properties in CD133-positive glioblastoma stem cells, suggesting that crosstalk between stem cell markers is required for CSC properties [58]. Although the high Sox2 level is associated with poor prognosis in many cancers including prostate [54], breast [56] and ovarian cancers [59], low Sox2 level in gastric cancer [60], and squamous cell lung cancer [61] was linked to poor prognosis. The reasons for the divergent results for Sox2 levels in different cancers remain to be investigated. Consequently, it is evident that there is a clear need for further investigations into the clinical implications of Sox2 expression, particularly how Sox2 levels influence tumor progression and patient survival.

In conclusion, we recognized that patients with CD133 and Sox2 overexpression had an unfavorable prognosis and poor clinical response to the current protocol of therapy suggesting the ineffectiveness of the current therapeutic modalities in these patients. Further investigations are needed to assess whether reversing the expression of these stem cell markers could be a novel therapeutic approach to prolong survival for this high-risk group of patients.

Footnotes

Conflict of interest

All of the authors declare that they have no conflict of interest regarding this paper.

References

Ferreira

W.A.

Araujo

M.D.

Anselmo

N.P.

Harada

M.L.

and Borges

B.D.

, Analysis of the methylation pattern of Sox2 and OCT4 genes in astrocytomas, Journal of Genetics and Genome Research 2 (2015), 1, ISSN: 2378-3648.

Zalata

K.R.

El-tantawy

D.A.

Abdel-Aziz

Ibraheim

A.W.M.

Halaka

A.H.

Gawish

H.H.

et al., Frequency of central nervous system tumors in delta region, Egypt, Indian J pathol Microbiol 54(2) (2011), 299–306.

Elsir

Edqvist

P.H.

Carlson

Ribom

Bergqvist

Ekman

et al., A study of embryonic stem cell-related proteins in human astrocytomas: Identification of Nanog as a predictor of survival, Int J Cancer 134 (2014), 1123–1131.

Dahlrot

R.H.

Hermansen

S.K.

Hansen

and Kristensen

B.W.

, What is the clinical value of cancer stem cell markers in gliomas? Int J Clin Exp Pathol 6 (2013), 334–348.

Chosdol

Misra

Puri

Srivastava

chattopadhyay

Sarkar

et al., Frequent loss of heterozygosity and altered expression of the candidate tumor suppressor gene FAT in human astrocytic tumors, BMC Cancer 9 (2009), 5. doi: 10.1186/1471-2407-9-5.

Hermansen

S.K.

Christensen

K.G.

Jensen

S.S.

and Kristensen

B.W.

, Inconsistent immunohistochemical expression patterns of four different CD133 antibody clones in glioblastoma, J Histochem Cytochem 59(4) (2011), 391–407.

Mak

A.B.

Blakely

K.M.

Williams

R.A.

Penttila

P.A.

Shukalyuk

A.L.

Osman

K.T.

et al., CD133 N-glycosylation processing contributes to cell-surface recognition of the primitive cell marker AC133, J Biol Chem 286(47) (2011), 41046–56.

Wang

Ahmad

Wang

et al., Expression of Sox2 and oct4 and their clinical significance in human non-small-cell lung cancer, Int J Mol Sci 13 (2012), 7663–7675.

Boumahdi

Driessens

Lapouge

Rorive

Nassar

Le Mercier

et al., Sox2 controls tumour initiation and cancer stem-cell functions in squamous-cell carcinoma, Nature 511 (2014), 246–250.

10.

Eschbacher

J.M.

Yeh

R.F.

Smirnov

Feuerstein

and Coons

, Sox2: A glioma-specific marker and a potential target for therapy, FASEB J 22(1) (2008), 504–504.

11.

Jesse

L.C.

Wilder

P.J.

Desler

and Rizzino

, Elevating SOX2 levels deleteriously affects the growth of medulloblastoma and glioblastoma cells, PLoS One 7 (2012), e44087.

12.

Louis

D.N.

Perry

Reifenberger

von Deimling

Figarella-Branger

Cavenee

W.K.

et al., The 2016 World Health Organization Classification of Tumors of the Central Nervous System: A summary, Acta Neuropathol 131(6) (2016), 803–820.

13.

Shin

J.H.

Lee

Y.S.

Hong

Y.K.

and Kang

C.S.

, Correlation between the prognostic value and the expression f the stem cell marker CD133 and isocitrate dehydrogenase1 in glioblastomas, J Neurooncol 115(3) (2013), 333–41.

14.

Balça-Silva

Matias

Dubois

L.G.

Carneiro

do Carmo

Girão

et al., The expression of connexins and SOX2 reflects the plasticity of glioma stem-like cells, Transl Oncol 10(4) (2017), 555–569.

15.

Khasraw

Ameratunga

M.S.

Gran

Wheeler

and Pavlakis

, Antiangiogenic therapy for high-grade glioma, Cochrane Database Syst Rev 22(9) (2014). doi: 10.1002/14651858.CD008218.pub3.

16.

Bexell

Svensson

and Bengzon

, Stem cell-based therapy for malignant glioma, Cancer Treat Rev 39(4) (2013), 358–65. doi: 10.1016/j.ctrv.201206.006.

17.

Wang

Zeng

Hanjun

L.I.

Zhang

and Wang

, Roles of sex-determining region Y-box 2 in cell pluripotency and tumor-related signaling pathways, Mol Clin Oncol 3(6) (2015), 1203–1207.

18.

Brescia

Richichi

and Pelicci

, Current strategies for identification of glioma stem cells: Adequate or unsatisfactory? J Oncol 2012 (2012), 10. Article ID 376894.

19.

Chen

McKay

R.M.

and Parada

L.F.

, Malignant glioma: Lessons from genomics, mouse models and stem cells, Cell 149 (2012), 36–47.

20.

Rasper

Schafer

Piontek

Teufel

Brockhoff

Ringel

et al., Aldehyde dehydrogenase 1 positive glioblastoma cells show brain tumor stem cell capacity, Neuro Oncol 12 (2010), 1024–33.

21.

Zhang

Song

Yang

Chen

Yang

Nestin et al. and CD133: Valuable stem cell-specific markers for determining clinical outcome of glioma patients, J Exp Clin Cancer Res 27 (2008), 85. doi: 10.1186/1756-9966-27-85.

22.

Melguizo

Prados

González

Ortiz

Concha

Alvarez

P.J.

et al., MGMT promoter methylation status and MGMT and CD133 immunohistochemical expression as prognostic markers in glioblastoma patients treated with temozolomide plus radiotherapy, J Transl Med 10 (2012), 250.

23.

McCrudden

C.M.

Yuen

H.F.

Lyu

Chan

K.W.

et al., CD133 in brain tumor: The prognostic factor, Oncotarget 8(7) (2017), 11144–11159.

24.

Dahlrot

R.H.

Hansen

Jensen

S.S.

Schroder

H.D.

Hjelmborg

and Kristensen

B.W.

, Clinical value of CD133 and nestin in patients with glioma: A population-based study, Int J Clin Exp Pathol 7(7) (2014), 3739–51.

25.

Sui

Y.P.

Jian

X.P.

L.I.

G.Z.

Liao

H.W.

Liu

Y.P.

et al., Prognostic value of cancer stem cell marker CD133 expression in esophageal carcinoma: A meta-analysis, Mol Clin Oncol 4(1) (2016), 77–82.

26.

Meng

Cui

Luo

and Xue

, Expression of the potential cancer stem cell markers CD133 and CD44 in medullary thyroid carcinoma: A ten-year follow-up and prognostic analysis, J Surg Oncol 113(2) (2016), 144–51.

27.

Zhao

and Zheng

, Prognostic value of cancer stem cell marker CD133 expression in pancreatic ductal adenocarcinoma (PDAC): A systematic review and meta-analysis, Int J Clin Exp Pathol 8(10) (2015), 12084–92.

28.

Zhou

Chen

Song

Tao

Yang

and Zuo

, Prognostic value of cancer stem cell marker CD133 in ovarian cancer: A meta-analysis, Int J Clin Exp Med 8(3) (2015), 3080–8.

29.

Angelastro

J.M.

and Lamé

M.W.

, Overexpression of Cd133 promotes drug resistance in C6 glioma cells, Molecular Cancer Research 8(8) (2010), 1105–15.

30.

Tamura

Aoyagi

Wakimoto

Ando

Nariai

Yamamoto

et al., Accumulation of CD133-positive glioma cells after high-dose irradiation by Gamma Knife surgery plus external beam radiation, J Neurosurgery 113(2) (2010), 310–8.

31.

Chang

C.J.

Hsu

C.C.

Yung

M.C.

Chen

K.Y.

Tzao

W.F.

et al., Enhanced radiosensitivity and radiation-induced apoptosis in glioma CD133-positive cells by knockdown of SirT1 expression, Biochem Biophys Res Commun 380(2) (2009), 236–242. doi: 10.1016/j.bbrc.2009.01.040.

32.

Song

W.S.

Yang

Y.P.

Huang

C.S.

K.H.

Liu

W.H.

W.W.

Sox2 et al., a stemness gene, regulates tumor-initiating and drug-resistant properties in CD133-positive glioblastoma stem cells, J Chin Med Assoc 79(10) (2016), 538–45.

33.

Bao

McLendon

R.E.

Hao

Shi

Hjelmeland

A.B.

et al., Glioma stem cells promote radioresistance by preferential activation of the DNA damage response, Nature 444 (2006), 756–60.

34.

Fouse

S.D.

Nakamura

J.L.

James

C.D.

and Costello

J.F.

, Response of primary glioblastoma cells to therapy is patient specific and independent of cancer stem cell phenotype, Neuro-Oncology 16(3) (2014), 361–371.

35.

Murat

Migliavacca

Gorlia

Lambiv

W.L.

Shay

Hamou

M.F

et al., Stem cell-related “self-renewal” signature and high epidermal growth factor receptor expression associated with resistance to concomitant chemoradiotherapy in glioblastoma, J Clin Oncol 26 (2008), 3015–3024.

36.

Arnold

Sarkar

Yram

M.A.

Polo

J.M.

Bronson

Sengupta

et al., Sox2+ adult stem and progenitor cells are important for tissue regeneration and survival of mice, Cell Stem 9 (2011), 317–329.

37.

Annovazzi

Mellai

Caldera

Valente

and Schiffer

, SOX2 expression and amplification in gliomas and glioma cell lines, Cancer Genomics and Proteomics 8(3) (2011), 139–148.

38.

Schmitz

Temme

Senner

Ebner

Schwind

Stevanovic

et al., Identification of SOX2 as a novel glioma-associated antigen and potential target for T cell-based immunotherapy, British Journal of Cancer 96 (2007), 1293–1301.

39.

Wang

Shi

et al., Associations between SOX2 and miR-200b expression with the clinicopathological characteristics and prognosis of patients with glioma, Experimental and Therapeutic Medicine (2015), 88–96.

40.

Gangemi

R.M.R.

Griffero

Marubbi

Perera

Capra

M.C.

Malatesta

et al., SOX2 silencing in glioblastoma tumor-initiating cells causes stop of proliferation and loss of tumorigenicity, Stem Cells 27(1) (2009), 40–48.

41.

Yang

Y.P.

Chien

Chiou

G.Y.

Cherng

J.Y.

Wang

M.L.

W.L.

et al., Inhibition of cancer stem cell-like properties and reduced chemoradioresistance of glioblastoma using microRNA145 with cationic polyurethane-short branch PEI, Biomaterials 33 (2012), 1462–1476.

42.

Alonso

M.M.

Diez-Valle

Manterola

Rubio

Liu

Cortes-Santiago

Genetic et al. and epigenetic modifications of Sox2 contribute to the invasive phenotype of malignant gliomas, PLoS One 6 (2011), e26740.

43.

Garros-Regulez

Aldaz

Arrizabalaga

Moncho-Amor

Carrasco-Garcia

Manterola

et al., mTOR inhibition decreases SOX2-SOX9 mediated glioma stem cell activity and temozolomide resistance, Expert Opin Ther Targets 20(4) (2016), 393–405. doi: 10.1517/14728222.2016.1151002.

44.

Fang

Yoon

J.G.

Shao

Hua

et al., The SOX2 response program in glioblastoma multiforme: An integrated ChIP-seq, expression microarray, and microRNA analysis, BMC Genomics 12 (2011), 11. https://doiorg/10.1186/1471-2164-12-11.

45.

Phi

J.H.

Park

S.H.

Kim

S.K.

Paek

S.H.

Kim

J.H.

Lee

Y.J.

et al., Sox2 expression in brain tumors: A reflection of the neuroglial differentiation pathway, Am J Surg Pathol 32 (2008), 103–112.

46.

Möllerström

Rydenhag

Andersson

Lebkuechner

Puschmann

T.B.

Chen

et al., Classification of subpopulations of cells within human primary brain tumors by single cell gene expression profiling, Neurochemical Research 40(2) (2015), 336–352.

47.

Chen

Huang

Chen

Wang

and Zhou

, The prognostic value of SOX2 expression in non-small cell lung cancer: A meta-analysis, PLoS One 8(8) (2013), e71140. https://doiorg/10.1371/journal.pone.0071140.

48.

González-Márquez

lorente

J.L.

Rodrigo

J.P.

García-Pedrero

J.M.

álvarez-Marcos

Suárez

et al., SOX2 expression in hypopharyngeal, laryngeal and sinonasal squamous cell carcinoma, Hum Pathol 45(4) (2014), 851–857.

49.

Rodriguez-Pinilla

S.M.

Sarrio

Moreno-Bueno

Rodriguez-Gil

Martinez

M.A.

et al., Sox2: a possible driver of the basal-like phenotype in sporadic breast cancer, Mod Pathol 20 (2007), 474–481.

50.

Sathyan

Zinn

P.O.

Marisetty

A.L.

Liu

Kamal

M.M.

Singh

S.K.

et al., Mir-21-Sox2 axis delineates glioblastoma subtypes with prognostic impact, J Neurosci 35 (2015), 15097–112. doi: 10.1523/JNEUROSCI.1265-15.2015.

51.

Galatro

T.F.

Uno

Oba-Shinjo

S.M.

Almeida

A.N.

Teixeira

M.J.

Rosemberg

et al., Differential expression of ID4 and its association with TP53 mutation, SOX2, SOX4 and OCT-4 expression levels, PLoS One 8(4) (2013), e61605. (ISSN: 1932-6203).

52.

Wan

Herold-Mende

Campos

Centner

F.S.

Dictus

Becker

et al., Association of stem cell-related markers and survival in astrocytic gliomas, Biomarkers 16 (2011), 136–143.

53.

Lemke

Weiler

Blaes

Wiestler

and Klein

, Primary glioblastoma cultures. Can profiling of stem cell markers predict radiotherapy sensitivity? Journal of Neurochemistry 131(2) (2014), 251–264.

54.

Jia

Zhang

Mou

Liu

et al., Sox2 promotes tumorigenesis and increases the anti-apoptotic property of human prostate cancer cell, Journal of Molecular Cell Biology 3(4) (2011), 230–238.

55.

Bareiss

P.M.

Paczulla

Wang

Schairer

Wiehr

Kohlhofer

et al., Sox2 expression associates with stem cell state in human ovarian carcinoma, Cancer Research 73(17) (2013), 5544–5555.

56.

Piva

Domenici

Iriondo

Rabano

Simoes

B.M.

Comaills

et al., Sox2 promotes tamoxifen resistance in breast cancer cells, EMBO Molecular Medicine 6 (2014), 66–79.

57.

Chou

M.Y.

F.W.

C.H.

and Yu

C.C.

, Sox2 expression involvement in the oncogenicity and radiochemoresistance of oral cancer stem cells, Oral Oncology 51(1) (2015), 31–39.

58.

Song

W.S.

Yang

Y.P.

Huang

C.S.

K.H.

Liu

W.H.

W.W.

Sox2 et al., a stemness gene, regulates tumor-initiating and drug-resistant properties in CD133-positive glioblastoma stem cells, Journal of the Chinese Medical Association 79(10) (2016), 538–545.

59.

Zhang

Chang

D.Y.

Mercado-Uribe

and Liu

, Sex-determining region Y-box 2 expression predicts poor prognosis in human ovarian carcinoma, Human Pathology 43 (2012), 1405–1412.

60.

Chen

Huang

Zhu

Chen

Huang

Zhang

et al., SOX2 inhibits metastasis in gastric cancer, Journal of Cancer Research and Clinical Oncology 142(6) (2016), 1221–1230.

61.

Wilbertz

Wagner

Petersen

Stiedl

A.C.

Scheble

V.J.

Maier

et al., SOX2 gene amplification and protein overexpression are associated with better outcome in squamous cell lung cancer, Modern Pathology 24 (2011), 944–953.

The clinicopathological significance of CD133 and Sox2 in astrocytic glioma

Abstract

BACKGROUND:

OBJECTIVES:

METHODS:

RESULTS:

CONCLUSIONS

Keywords

1. Introduction

2. Materials and methods

2.1 Patients and tumor specimens

2.2 Immunohistochemistry

2.3 Assessment of the immunohistochemistry

2.3.2 Sox2 scoring

2.4 Statistics

Table 1 Clinicopathological features of 40 astrocytic glioma patients

3.1 Patients’ characteristics

3.2 The association of CD133 and Sox2 expression with the clinicopathological features

Table 2 Relations between clinicopathological features and CD133 expression

4. Discussion

Footnotes

Conflict of interest

References

Table 1
Clinicopathological features of 40 astrocytic glioma patients

Table 2
Relations between clinicopathological features and CD133 expression