Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Sal-like protein 4 (SALL4) is recognized as a potential biomarker for assessing cancer prognosis. Many experiments have been done to explore the aberrant expression of SALL4 in cancer patients.

OBJECTIVE:

To explore the association between SALL4 expression and cancer prognosis.

METHODS:

A systematically comprehensive search was performed through PubMed, Web of Science, and CNKI. The prognostic value of SALL4 expression in cancer patients was evaluated by pooled hazard ratios (HRs) and 95% confidence intervals (CIs).

RESULTS:

The overexpression of SALL4 was identified to be significantly associated with a poor prognosis in cancer patients (pooled HR: 2.02, 95% CI: 1.67–2.46). This association was also detected in digestive system neoplasms subgroup (pooled HR: 1.91, 95% CI: 1.52–2.39) and reproductive system neoplasms subgroup (pooled HR: 1.95, 95% CI: 1.17–3.26). In the geography subgroup analysis, a statistical association was confirmed in the Chinese subgroup (pooled HR: 1.982, 95% CI: 1.526–2.576). In the respect of disease progression, there was a certain degree of relationship between higher expression of SALL4 and recurrence-free survival (RFS) (pooled HR: 1.617, 95% CI: 1.190–2.196).

CONCLUSION:

SALL4 is a promising prognostic biomarker for cancer, and is appropriate for the assessment of cancer prognosis in the Chinese people.

Keywords

SALL4 cancer biomarker prognosis meta-analysis

1. Introduction

SALL4, as a zinc-finger transcription factor, was originally considered as a homolog of the Drosophila homeotic gene spalt [1]. Afterwards, it deemed to play an essential role in embryonic development and self-renewal of embryonic stem cells through interactions with OCT3/4, SOX2, and NANOG [2, 3, 4, 5]. The functional loss of SALL4 may result in dysplasia of multiple organs and abortion or embryonic death in mice [2, 6]. In humans, SALL4 gene mutations can lead to an autosomal dominant disease called Okihiro syndrome (Duane radial ray syndrome), which causes multiple organ dysfunction [7, 8, 9, 10]. As tissues and organs mature, SALL4 expression gradually reduced [11]. However, accumulating evidence indicates that SALL4 expression is reactivated in numerous human malignancies. It was the first time that SALL4 was confirmed as an oncogene in leukemia [12]. Consecutively, the overexpression of SALL4 has been detected in a variety of solid cancers and indicated a poor clinical prognostic value in a few of them, such as germ cell tumor [13, 14], hepatocellular carcinoma [15], colorectal cancer [16], gastric cancer [17], lung cancer [18] and breast cancer [19]. However, some researcher drew the inconclusive conclusion [20]. Thus, whether higher expression of SALL4 predicts poor cancer prognosis remains confused. In view of the relatively small sample sizes of studies and the different results, we conducted this meta-analysis of all available studies in order to acquire more reliable conclusion about the relationship between SALL4 expression and the prognosis influence in cancer patients.

2. Materials and methods

2.1 Literature search

A systematic literature search was carried out by means of the databases PubMed, Web of Science and CNKI for studies in humans of the association between SALL4 expression and cancer prognosis. The search strategy used the terms: ([sal-like protein 4 or SALL4] AND [carcinoma or tumor or cancer] AND [prognosis or survival or outcome]) covering all relevant studies published up to March 2016. Meanwhile, the population, sample size, time period or types of clinical trials on the retrieved studies were not limited, and only studies in English or Chinese were included. Reference lists from relevant articles were inspected manually to further find additional eligible studies. Two investigators performed the search independently. When some overlapping studies were checked, we chose a better quality study or one with more detailed data.

2.2 Inclusion criteria

To be eligible for inclusion, all the studies have to meet the following criteria: (1) definite diagnosis of cancer in humans, (2) trials published as a full paper in English or Chinese literature, (3) evaluating the association between SALL4 and cancer prognosis, (4) SALL4 quantity (protein or mRNA) analysis of the primary tumor (not in metastatic tumor or in tumor adjacent tissues) or in the serum, (5) sufficient patient data for eligible HRs with 95% CIs or other available data for HR with a 95% CI (including data extracted from Kaplan-Meier curves). The major reasons for exclusion of studies were: (1) overlapping data; (2) abstract, comment, editorials letters, and review; (3) studies without detailed data. The study selection flow diagram was shown in Fig. 1.

Figure 1.

Flow chart of the selection of publications included in the meta-analysis.

Table 1

Main characteristics of studies included in the meta-analysis

First	Year	Case	Sample	Mean	Cancer type	Survival	Source	Follow-up	NOS
author		nationality	size	age		analysis	of HR	months	score
He	2012	China	90	NA	Ovarian malignant germ cell tumor	OS/DFS	Reported	160.23	6
Yong1	2013	China	228	NA	Hepatocellular Carcinoma	OS/RFS	Reported	24	7
Yong2	2013	Singapore	79	56	Hepatocellular Carcinoma	OS/RFS	Reported	24	8
Oikawa	2013	China/USA	110	NA	Hepatocellular carcinoma	OS	SC	36	6
Zeng	2014	Japan	144	NA	Hepatocellular carcinoma	RFS	SC	12	6
Shibahara	2014	Japan	337	64.6	Hepatocellular carcinoma	OS/DFS	Reported	48	8
Liu	2014	China	84	53	Small cell lung cancer	OS	SC	72	6
Han	2014	China	56	NA	Hepatocellular Carcinoma	OS	SC	24	6
Park	2015	Korea	190	58.1 $\pm$ 11.8	Hepatocellular carcinoma	OS	Reported	133	7
Yue	2015	China	160	51	Breast invasive ductal carcinoma	OS	SC	144	5
Lei	2015	China	80	NA	Endometrial cancer	OS	SC	NA	6
Tanaka	2015	Japan	90	62.0 $\pm$ 11.2	Combined hepatocellular carci- noma and cholangiocarcinoma	OS/DFS	SC	NA	7
Zhang	2015	China	54	NA	Glioma	OS	Reported	96	7
Kilic1	2016	Germany	152	NA	Esophageal squamous cell carci- noma	OS	SC	NA	7
Kilic2	2016	Germany	115	NA	Esophageal adenocarcinoma	RFS	SC	NA	7
Yang	2016	China	91	NA	Serous ovarian carcinoma	OS/PFS	Reported	72	8

OS, overall survival; DFS, disease-free survival; RFS, recurrence-free survival; PFS, progression-free survival; NA, not available; SC, survival curve; HR, hazard ratio; NOS score, Newcastle-Ottawa Scale score.

2.3 Data extraction

Two investigators extracted the data independently according to the standard approach [21]. In case of discordant studies, the third investigator was invited to further review the original data until an agreement was reached. We collected the following data from each eligible publication: first author’s name, publication year, case nationality, sample size, mean age, cancer type, follow-up months, detecting method (immunohistochemistry (IHC) or tissue microarray (TMA) or enzyme-linked immuno sorbent assay (ELISA)), specimen, and cut-off value, respectively. Most importantly, the HR with the corresponding 95% CI for overall survival (OS), disease-free survival (DFS), recurrence-free survival (RFS) or progression-free survival (PFS) were extracted from every included study. Since the study only provided RR (risk ratio) with 95% CI, the RR is approximately considered as HR [22]. When the HRs with 95% CIs were described in text or tables, we acquired them directly [22, 23, 24, 25, 26, 27]. Otherwise, we obtained the necessary data from the only provided K-M survival curves using Engauge Digitizer 4.1 [28] by two independent researchers [29, 30, 31, 32, 33, 34, 35, 36]. For studies providing more than one HR, the HRs from multivariate models with the most completed adjustment for confounding factors were extracted for further analysis.

2.4 Study quality assessment

The quality assessment of the studies included in present meta-analysis was independently evaluated by two investigators according to the Newcastle-Ottawa quality assessment scale (NOS) [37]. Each study was assessed on the base of three broad perspectives: selection, comparability and exposure with a score ranging from 0 to 9. Studies with a score of 7 or greater were considered high quality studies. Discrepancies were adjudicated by discussion or through consultation with a third investigator.

2.5 Statistical analysis

HR with a 95% CI was assessed for the relationship between SALL4 expression and cancer prognosis (OS and DFS/RFS/PFS, respectively). An observed HR $>$ 1 predicted a poor prognosis in patients with higher SALL4 expression. The heterogeneity was calculated using a Chisquare based Q-test [38], with $I^{2}>$ 50% or $P<$ 0.1 considered statistically significant [39]. If there was no significant heterogeneity, the fixed effects model (Mantel-Haenszel method) was utilized. Otherwise, we used the random effects model (the Der Simonian and Laird method) [40]. Thereafter, subgroup analyses were performed to explore the impact of some variables on the pooled HRs. Sensitivity analysis was conducted to identify the effect of the individual data on the summary HRs. Publication bias was determined using Egger’s test and Begg’s funnel plots [41]. All statistical analyses were conducted with STATA software version 12.0 (STATA Corporation, College Station, TX, USA) and $P<$ 0.05 was considered significant.

Table 2
HRs and 95% CIs for patient survival in association with SALL4 expression in enrolled studies

First Detecting Specimen Cut-off Sample size HR (95% CI)

author method value Higher Lower OS DFS/RFS/PFS

expression expression

He IHC Tumor Score $\geqslant$ 5 28 62 2.13 (0.592–7.667)M 3.312 (1.85–7.146)M

Yong1 TMA Tumor Median 114 114 1.52 (1.00–2.32)M 1.67 (1.11–2.51)M

Yong2 IHC Tumor Score $\geqslant$ 1 45 34 2.87 (1.09–7.52)M 2.7 (0.41–18.72)M

Oikawa TMA Tumor Median 55 55 1.51 (0.75–3.04)U* NA

Zeng IHC Tumor $>$ 5% cells stained 101 43 NA 2.74 (1.10–6.80)U*

Shibahara IHC Tumor $>$ 2% cells stained 47 290 1.155 (0.591–2.555)U 0.843 (0.457–1.554)M

Liu IHC Tumor Score $\geqslant$ 3 59 25 3.07 (1.29–7.28)U* NA

Han ELISA Serum $\geqslant$ 130.43 pg/mL 29 27 2.16 (1.03–4.49)U* NA

Park IHC Tumor $>$ 10% cells stained 39 151 3.556 (1.511–8.367)M NA

Yue IHC Tumor Score $>$ 5 103 57 1.85 (0.74–4.62)U* NA

Lei IHC Tumor Score $\geqslant$ 4 56 24 1.16 (0.34–3.93)U* NA

Tanaka IHC Tumor $\geqslant 5\%$ cells stained 8 82 2.15 (0.94–4.95)U* 1.05 (0.46– 2.43)U*

Zhang IHC Tumor Score $\geqslant$ 2 39 15 6.416 (2.053–20.052)M NA

Kilic1 IHC Tumor Score $\geqslant$ weak 14 138 2.38 (1.48–3.82)U* NA

Kilic2 IHC Tumor Score $\geqslant$ weak 22 93 NA 1.19 (0.68–2.09)U*

Yang IHC Tumor Score $\geqslant$ 2 53 38 2.188 (1.242–4.379)M 2.961 (1.614–4.721)M

The source of HR (hazard ratio) and 95% CI (confidence interval) is represented as derived from univariate analysis (U) or multivariate analysis (M). *HR and 95% CI calculated from K-M survival curves. OS, overall survival; DFS, disease-free survival; RFS, recurrence-free survival; PFS, progression-free survival; HR, HR (higher vs lower); NA, not available; IHC, immunohistochemistry; TMA, tissue microarray; ELISA, enzyme-linked immuno sorbent assay.

3. Results

3.1 Characteristics of the studies

The flow chart describing the process of study selection was presented in Fig. 1. From a total of 125 potentially relevant articles through searching from the databases, we excluded 58 duplicates and 13 irrelevant articles. After a rough review of the titles and abstracts, 33 studies were excluded because of no relationship between SALL4 expression and cancer prognosis. In the remaining 27 records, with a systematical review of the full texts, another 13 articles were excluded for lacking of insufficient data. Eventually, 14 eligible articles (16 studies) containing 2,060 patients were included in this meta-analysis [22, 23, 24, 25, 26, 27, 29, 30, 31, 32, 33, 34, 35, 36]. Yong and Kilic’s studies included two different survival analyses due to two different case nationality and two different pathological types, separately [24, 36].

The main characteristics of all the eligible studies are summarized in Tables 1 and 2. Of the 16 studies, 14 studies were extracted for OS [22, 23, 24, 25, 26, 27, 29, 31, 32, 33, 34, 35, 36] while 8 studies were obtained for DFS/RFS/PFS [23, 24, 25, 27, 30, 35, 36]. Except Oikawa’s study [29] based on both Chinese and White, 13 studies (1683 patients: 81.7%) were conducted in Asian area, and only two studies (267 patients: 13.0%) were performed in European area. The malignant tumors assessed in these studies included ovarian malignant germ cell tumor [23], hepatocellular carcinoma (HCC) [24, 25, 26, 29, 30, 32], combined hepatocellular carcinoma and cholangiocarcinoma, [35] small cell lung cancer, [31] breast cancer, [33] endometrial cancer [34], glioma [22], esophagus cancer [36] and serous ovarian carcinoma [27]. IHC was used to detect the expression of SALL in 13 studies, except that Yong1 and Oikawa’s methods were performed by TMA [24, 29] and Han’s were detected via ELISA [32].

3.2 Evidence synthesis

3.2.1 SALL4 expression and OS

Fourteen studies with a total of 1,801 patients which provided survival results in the form of OS were enrolled in this meta-analysis. Since the heterogeneity was not statistically significant ( $I^{2}=$ 0.8%, $P=$ 0.440), the fixed model was applied to pool HRs. Our result showed that overexpression of SALL4 was significantly associated with poor OS in multiple carcinomas, with the pooled HR of 2.02 (95% CI: 1.67–2.46, $P<$ 0.001) (Fig. 2).

Figure 2.

Forest plot of HR for the association between higher SALL4 expression and OS in patients with malignancies.

To explore the prognostic role of SALL4 in different carcinomas, we divided 14 studies providing results of OS into subgroups by human body system. The results suggested that higher SALL4 expression was an infaust prognostic marker in patient with digestive system neoplasms (HR: 1.91, 95% CI: 1.52–2.39, $P<$ 0.001), reproductive system neoplasms (HR: 1.95, 95% CI: 1.17–3.26, $P=$ 0.011) and others (HR: 3.04, 95% CI: 1.75–5.27, $P<$ 0.001), and no heterogeneity was discovered within these groups (Fig. 3).

Table 3

Main results of meta-analysis

Categories	Studies	Patients	SALL4 $+$	HRs	95% CI	Heterogeneity		$P$
						I-Square	$P_{h}$
Survival type
OS*	14	1801	689	2.02	1.67–2.46	0.8%	0.440	$<$ 0.001
DFS	3	517	83	1.436	0.599–3.443	78.6%	0.009	0.418
RFS	4	436	282	1.617	1.190–2.196	0.0%	0.434	0.002
PFS	1	91	53	2.961	1.614–4.721	NA	NA	$<$ 0.001
System*
Digestive system neoplasms	8	1242	351	1.91	1.52–2.39	1.4%	0.419	$<$ 0.001
Reproductive system neoplasms	3	261	137	1.95	1.17–3.26	0.0%	0.657	0.011
Others	3	298	201	3.04	1.75–5.27	28.1%	0.249	$<$ 0.001
Geography*
Chinese	8	843	481	1.982	1.526–2.576	6.8%	0.378	$<$ 0.001
Non-Chinese	5	848	153	2.212	1.617–3.025	12.8%	0.333	$<$ 0.001
Methods*
IHC	11	1407	491	2.29	1.80–2.91	0.0%	0.473	$<$ 0.001
TMA	2	338	169	1.52	1.06–2.18	0.0%	0.987	0.023
ELISA	1	56	29	2.16	1.03–4.49	NA	NA	0.04

PFS, progression-free survival; SALL4 $+$ , SALL4 positive patients number; HR, hazard ratio; CI, confidence interval; $P_{h}$ , $P$ Heterogeneity; OS, overall survival; DFS, disease-free survival; RFS, recurrence-free survival; IHC, immunohistochemistry; TMA, tissue microarray; ELISA, enzyme-linked immuno sorbent assay; NA, not available. The marker (*) means calculation with overall survival data.

Figure 3.

Meta-analysis (Forest plot) of the eligible studies assessing SALL4 expression and OS in cancer patients stratified by human body system.

We also conducted subgroup analysis by geography and detecting methods. The results indicated that there were significant correlations between SALL4 overexpression and poor OS in studies with Chinese (HR: 1.982, 95% CI: 1.526–2.576, $P<$ 0.001) and non-Chinese (HR: 2.212, 95% CI: 1.617–3.025, $P<$ 0.001). Both of the two groups has no obvious heterogeneity. Finally, for the methods subgroup, a statistical relationship was observed in the IHC subgroup (pooled HR: 2.29, 95% CI: 1.80–2.91, $P<$ 0.001), which was the most critical method for detecting SALL4 expression (Table 3).

3.2.2 SALL4 expression and DFS/RFS/PFS

Three studies for DFS, four studies for RFS and only one study for PFS were performed separately in our meta-analysis due to different survival type owning different definition. With a random-effects model due to significant heterogeneity ( $I^{2}=$ 78.6%, $P=$ 0.009), our results failed to prove any significant association between SALL4 expression and DFS (HR: 1.436, 95% CI: 0.599–3.443, $P=$ 0.418) (Table 3). However, there may be a substantial link between higher expression of SALL4 and RFS (HR: 1.617, 95% CI: 1.190–2.196, $P=$ 0.002). Subgroup analysis, publication bias and sensitivity analysis were not appropriate in analysis of the relationship between SALL4 expression and DFS/ RFS/PFS on account of the limited number of enrolled studies.

3.3 Sensitivity analysis

To assess the robustness of SALL4 expression and patient overall survival, a sensitivity analysis was adopted through excluding the enrolled studies one by one and analyzing the influence and homogeneity of the remaining studies. Both random effects model and fixed effects model were used and the two models did not cause any significant effect on the results. The sensitivity analysis results demonstrated that pooled HRs was not materially altered, indicating that no significant changes were existed in the HRs when ruling out any of the studies.

3.4 Publication bias

Begg’s funnel plot and Egger’s test were used to evaluate the publication bias of all included studies providing OS. The graphical funnel plots showed no evidence of asymmetry. Then, Egger’s test was applied to provide statistical evidence of funnel plot symmetry. The $P$ value ( $P=$ 0.230) still did not reveal any evidence of publication bias in the overall survival meta-analysis (Fig. 4).

Figure 4.

Begg’s funnel plot of SALL4 expression and OS in cancer patients.

4. Discussion

Cancer remains the most public health burden, attributed to multiaspect reasons like stress, lifestyle, environment, psychological/personality, and many health -related factors [42]. It is of great importance to look for reliable and favorable prognostic biomarkers for cancer patients and further provide available information for clinical decision making. SALL4 is considered to be a valuable and specific cancer biomarker. SALL4 expression is reported in numerous malignancies and correlates with overall survival and disease progression [16, 17]. In an early clinical stage of breast cancer, SALL4 is expressed at a high level, indicating that SALL4 may be conducive to breast cancer screening [19]. In human HCC, SALL4 expression is restored in tumor tissues as well as serum levels. According to both tumor recurrence and overall survival rate, HCC patients with higher SALL4 expression are significantly associated with poor prognosis, suggesting that SALL4 is a novel and independent prognosis biomarker for HCC patients [15, 29, 32, 43]. Moreover, the level of SALL4 expression is positively correlated with worse patient survival and disease characteristics such as invasion and metastasis in endometrial cancer [43], colorectal carcinoma [16, 44], esophageal carcinoma [45], and gastric cancer [17, 44]. Thus, to address the prognostic value of SALL4 expression, we conducted this meta-analysis.

To our knowledge, this is the first time that a meta-analysis has been performed to focus on the association between elevated expression of SALL4 and the prognosis of patients with cancers. An aggregate amount of 16 eligible studies, including 2,060 cases, were selected and analyzed in the present meta-analysis. The results revealed that higher SALL4 expression was significantly associated with a poor prognosis (HR 2.02, 95% CI: 1.67–2.46, $P<$ 0.001). This suggests that SALL4 can be used for predicting cancer prognosis.

In order to seek for more detailed conclusions, we stratified the enrolled studies for further analysis. Eight studies were enrolled in the digestive system neoplasms subgroup (pooled HR 1.91, 95% CI 1.52–2.39, $P<$ 0.001) and three in the reproductive system neoplasms subgroup (pooled HR 1.95, 95% CI 1.17–3.26, $P=$ 0.011). Both systems of cancer displayed a significantly decreased survival rate in patients with higher SALL4 expression. Thus, SALL4 could be considered as a novel prognostic marker for different types of carcinomas, especially for tumors from digestive and reproductive system. To determine the prognostic role of SALL4 in different regions, we conducted subgroup analysis by geography. Significant association was also found in Chinese patients (HR 1.982, 95% CI 1.526–2.576, $P<$ 0.001), suggesting SALL4 had an important prognostic value in Chinese ethnic groups. When in view of detecting methods, we found that the IHC subgroup (HR 2.29, 95% CI 1.80–2.91, $P<$ 0.001) was the most effective methods for detecting the expression of SALL4 in cancer patients. Although a significant relationship observed in TMA and ELISA subgroup, it is hard to draw conclusions for only three studies enrolled.

Concerning DFS in this meta-analysis, however, we failed to explore any significant association between SALL4 expression and DFS (HR 1.436, 95% CI 0.599–3.443, $P=$ 0.418) with significant heterogeneity ( $I^{2}=$ 78.6%, $P=$ 0.009). Although a significant result for RFS, considering the small sample size (three studies for DFS, four for RFS, only one for PFS), it may be too early to come to a conclusion and more researches are requisite to strengthen our conclusions.

Although our study is the first meta-analysis about the association between SALL4 expression and patient survival, some limitations should be taken into account. Firstly, our meta-analysis only contained a total of 16 studies, thus the results might be reached by accident because sample error of eligible studies could result in insufficient statistical power. Secondly, our meta-analysis did not encompass all human tumor types. Although SALL4 is a promising biomarker, the correlation still need further research and large-scale studies are required for further credible results. Thirdly, since most of the enrolled studies used IHC to detect SALL4 levels, a technique bias in different studies may exist for various dyeing operation, antibody concentration and cut-off value of different tissues. Fourthly, because the survival data were not presented directly in some included studies, we had to extract HRs and 95% CIs from survival curves and these data might be less reliable than those given directly. Finally, in spite of no significant difference detected in the results of sensitivity analysis and publication bias assay, publication bias cannot be utterly eliminated.

In conclusion, there was a positive correlation between SALL4 overexpression and poor prognosis of patients with cancers from the present meta-analysis. SALL4, therefore, may serve as a promising prognostic marker and an emerging potential therapeutic target for cancer patients. In order to intensify our conclusion, it is strongly suggested that more standardized randomized controlled trials (RCTs) and prospective studies with high quality should be carried out to dig into the molecular mechanisms of SALL4 and the relationship between higher SALL4 expression and prognosis for cancer patients.

Abbreviations

SALL4: sal-like protein 4,

CNKI: China National Knowledge Infrastructure,

HR: hazard ratio,

CI: confidence interval,

RFS: recurrence-free survival,

DFS: disease-free survival, OS: overall survival,

PFS: progression-free survival,

RR: risk ratio,

IHC: immunohistochemistry,

TMA: tissue microarray,

ELISA: enzyme-linked immuno sorbent assay,

NOS: Newcastle-Ottawa quality assessment scale,

HCC: hepatocellular carcinoma,

RCTs: randomized controlled trials.

Footnotes

Acknowledgments

This study was funded by the National Natural Science Foundation of China (81272470).

Conflict of interest

The authors declare that they have no conflict of interests.

References

de Celis

J.F.

and Barrio

, Regulation and function of Spalt proteins during animal development, Int J Dev Biol 53 (2009), 1385–1398.

Elling

Klasen

Eisenberger

Anlag

and Treier

, Murine inner cell mass-derived lineages depend on Sall4 function, Proc Natl Acad Sci U S A 103 (2006), 16319–16324.

Lim

C.Y.

Tam

W.L.

Zhang

Ang

H.S.

Jia

Lipovich

H.H.

Wei

C.L.

Sung

W.K.

Robson

Yang

and Lim

, Sall4 regulates distinct transcription circuitries in different blastocyst-derived stem cell lineages, Cell Stem Cell 3 (2008), 543–554.

Rao

Zhen

Roumiantsev

McDonald

L.T.

Yuan

G.C.

and Orkin

S.H.

, Differential roles of Sall4 isoforms in embryonic stem cell pluripotency, Mol Cell Biol 30 (2010), 5364–5380.

Yang

Gao

Chai

and Ma

, A novel SALL4/OCT4 transcriptional feedback network for pluripotency of embryonic stem cells, PLoS One 5 (2010), e10766.

Sakaki-Yumoto

Kobayashi

Sato

Fujimura

Matsumoto

Takasato

Kodama

Aburatani

Asashima

Yoshida

and Nishinakamura

, The murine homolog of SALL4, a causative gene in Okihiro syndrome, is essential for embryonic stem cell proliferation, and cooperates with Sall1 in anorectal, heart, brain and kidney development, Development 133 (2006), 3005–3013.

Kohlhase

and Holmes

L.B.

, Mutations in SALL4 in malformed father and daughter postulated previously due to reflect mutagenesis by thalidomide, Birth Defects Res A Clin Mol Teratol 70 (2004), 550–551.

Kohlhase

Chitayat

Kotzot

Ceylaner

Froster

U.G.

Fuchs

Montgomery

and Rosler

, SALL4 mutations in Okihiro syndrome (Duane-radial ray syndrome), acro-renal-ocular syndrome, and related disorders, Hum Mutat 26 (2005), 176–183.

Kohlhase

Heinrich

Schubert

Liebers

Kispert

Laccone

Turnpenny

Winter

R.M.

and Reardon

, Okihiro syndrome is caused by SALL4 mutations, Hum Mol Genet 11 (2002), 2979–2987.

10.

Al-Baradie

Yamada

St Hilaire

Chan

W.M.

Andrews

McIntosh

Nakano

Martonyi

E.J.

Raymond

W.R.

Okumura

Okihiro

M.M.

and Engle

E.C.

, Duane radial ray syndrome (Okihiro syndrome) maps to 20q13 and results from mutations in SALL4, a new member of the SAL family, Am J Hum Genet 71 (2002), 1195–1199.

11.

Zhang

Yuan

Zhu

Qian

and Xu

, SALL4: An emerging cancer biomarker and target, Cancer Lett 357 (2015), 55–62.

12.

Cui

Yang

Amin

H.M.

Lai

Ritz

Krause

D.S.

and Chai

, SALL4, a novel oncogene, is constitutively expressed in human acute myeloid leukemia (AML) and induces AML in transgenic mice, Blood 108 (2006), 2726–2735.

13.

Cao

Guo

C.C.

Allan

R.W.

and Humphrey

P.A.

, SALL4 is a novel diagnostic marker for testicular germ cell tumors, Am J Surg Pathol 33 (2009), 1065–1077.

14.

Nastaly

Ruf

Becker

Bednarz-Knoll

Stoupiec

Kavsur

Isbarn

Matthies

Wagner

Hoppner

Fisch

Bokemeyer

Ahyai

Honecker

Riethdorf

and Pantel

, Circulating tumor cells in patients with testicular germ cell tumors, Clin Cancer Res 20 (2014), 3830–3841.

15.

Masuda

Suzuki

and Izpisua Belmonte

J.C.

, Oncofetal gene SALL4 in aggressive hepatocellular carcinoma, N Engl J Med 369 (2013), 1171.

16.

Ardalan Khales

Abbaszadegan

M.R.

Abdollahi

Raeisossadati

Tousi

M.F.

and Forghanifard

M.M.

, SALL4 as a new biomarker for early colorectal cancers, J Cancer Res Clin Oncol 141 (2015), 229–235.

17.

Zhang

Wang

Zhang

Yang

Cai

Yan

Mao

Zhu

Shao

Qian

and Xu

, SALL4, a novel marker for human gastric carcinogenesis and metastasis, Oncogene 33 (2014), 5491–5500.

18.

Gautam

A.K.

Wang

Zeng

Wang

Wei

Huang

Luo

and Mo

, Expression and clinical significance of SALL4 and LGR5 in patients with lung cancer, Oncol Lett 10 (2015), 3629–3634.

19.

Kobayashi

Kuribayshi

Tanaka

and Watanabe

, SALL4 is essential for cancer cell proliferation and is overexpressed at early clinical stages in breast cancer, Int J Oncol 38 (2011), 933–939.

20.

Liu

T.C.

Vachharajani

Chapman

W.C.

and Brunt

E.M.

, SALL4 immunoreactivity predicts prognosis in Western hepatocellular carcinoma patients but is a rare event: A study of 236 cases, Am J Surg Pathol 38 (2014), 966–972.

21.

Moher

Liberati

Tetzlaff

Altman

D.G.

and Group

, Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement, J Clin Epidemiol 62 (2009), 1006–1012.

22.

Zhang

Yan

Jiang

Cui

Zou

Qian

Luo

and Wu

, The expression of SALL4 in patients with gliomas: High level of SALL4 expression is correlated with poor outcome, J Neurooncol 121 (2015), 261–268.

23.

Ding

Wang

J.H.

Zhou

Y.H.

Huang

Jia

W.H.

Zeng

Yun

J.P.

Luo

R.Z.

and Zheng

, Overexpression of karyopherin 2 in human ovarian malignant germ cell tumor correlates with poor prognosis, PLoS One 7 (2012), e42992.

24.

Yong

K.J.

Gao

Lim

J.S.

Yan

Yang

Dimitrov

Kawasaki

Ong

C.W.

Wong

K.F.

Lee

Ravikumar

Srivastava

Tian

Poon

R.T.

Fan

S.T.

Luk

J.M.

Dan

Y.Y.

Salto-Tellez

Chai

and Tenen

D.G.

, Oncofetal gene SALL4 in aggressive hepatocellular carcinoma, N Engl J Med 368 (2013), 2266–2276.

25.

Shibahara

Ando

Hayashi

Sakamoto

Hesegawa

Kokudo

and Fukayama

, Clinicopathologic characteristics of SALL4-immunopositive hepatocellular carcinoma, Springerplus 3 (2014), 721.

26.

Park

Lee

Seo

A.N.

Cho

J.Y.

Choi

Y.R.

Yoon

Y.S.

Han

H.S.

Park

Y.N.

and Kim

, SALL4 expression in hepatocellular carcinomas is associated with EpCAM-positivity and a poor prognosis, J Pathol Transl Med 49 (2015), 373–381.

27.

Yang

Xie

and Ding

, SALL4 is a marker of poor prognosis in serous ovarian carcinoma promoting invasion and metastasis, Oncol Rep 35 (2016), 1796–1806.

28.

Tierney

J.F.

Stewart

L.A.

Ghersi

Burdett

and Sydes

M.R.

, Practical methods for incorporating summary time-to-event data into meta-analysis, Trials 8 (2007), 16.

29.

Oikawa

Kamiya

Zeniya

Chikada

Hyuck

A.D.

Yamazaki

Wauthier

Tajiri

Miller

L.D.

Wang

X.W.

Reid

L.M.

and Nakauchi

, Sal-like protein 4 (SALL4), a stem cell biomarker in liver cancers, Hepatology 57 (2013), 1469–1483.

30.

Zeng

S.S.

Yamashita

Kondo

Nio

Hayashi

Hara

Nomura

Yoshida

Hayashi

Oishi

Ikeda

Honda

and Kaneko

, The transcription factor SALL4 regulates stemness of EpCAM-positive hepatocellular carcinoma, J Hepatol 60 (2014), 127–134.

31.

Liu

Bai

Wang

and Guo

, Role of SALL4 in regulating multi-drug resistance of small cell lung cancer and its clinical significance, Zhonghua Bing Li Xue Za Zhi 43 (2014), 604–608.

32.

Han

S.X.

Wang

J.L.

Guo

X.J.

C.C.

Ying

J.L.

Zhang

Y.Y.

Zhao

and Zhu

, Serum SALL4 is a novel prognosis biomarker with tumor recurrence and poor survival of patients in hepatocellular carcinoma, J Immunol Res 2014 (2014), 262385.

33.

Yue

Xiao

Yang

Liu

Zhang

Shi

Zhou

Geng

Ning

and Zhang

, High cytoplasmic expression of SALL4 predicts a malignant phenotype and poor prognosis of breast invasive ductal carcinoma, Neoplasma 62 (2015), 980–988.

34.

Liu

Zhang

Yang

Fang

and Xi

, SALL4 as an epithelial-mesenchymal transition and drug resistance inducer through the regulation of c-Myc in endometrial cancer, PLoS One 10 (2015), e0138515.

35.

Tanaka

Aishima

Kohashi

Okumura

Wang

Hida

Kotoh

Shirabe

Maehara

Takayanagi

and Oda

, Spalt-like transcription factor 4 immunopositivity is associated with epithelial cell adhesion molecule expression in combined hepatocellular carcinoma and cholangiocarcinoma, Histopathology (2015).

36.

Kilic

Tennstedt

Hogner

Lebok

Sauter

Bokemeyer

Izbicki

J.R.

and Wilczak

, The zinc-finger transcription factor SALL4 is frequently expressed in human cancers: Association with clinical outcome in squamous cell carcinoma but not in adenocarcinoma of the esophagus, Virchows Arch (2016).

37.

Stang

, Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses, Eur J Epidemiol 25 (2010), 603–605.

38.

Higgins

J.P.

and Thompson

S.G.

, Quantifying heterogeneity in a meta-analysis, Stat Med 21 (2002), 1539–1558.

39.

Hedges

L.V.

and Pigott

T.D.

, The power of statistical tests in meta-analysis, Psychol Methods 6 (2001), 203–217.

40.

DerSimonian

and Laird

, Meta-analysis in clinical trials, Control Clin Trials 7 (1986), 177–188.

41.

Egger

Davey Smith

Schneider

and Minder

, Bias in meta-analysis detected by a simple, graphical test, BMJ 315 (1997), 629–634.

42.

Zhang

Fielding

Soong

Chan

K.K.

Tsang

Lee

Sze

W.K.

Tin

and Lam

W.W.

, Illness perceptions among cancer survivors, Support Care Cancer 24 (2016), 1295–1304.

43.

Jiao

Yong

K.J.

Wang

Gao

Yan

Srivastava

Lim

G.S.

Tang

Yang

Tenen

D.G.

and Chai

, SALL4 is a new target in endometrial cancer, Oncogene 34 (2015), 63–72.

44.

Forghanifard

M.M.

Moghbeli

Raeisossadati

Tavassoli

Mallak

A.J.

Boroumand-Noughabi

and Abbaszadegan

M.R.

, Role of SALL4 in the progression and metastasis of colorectal cancer, J Biomed Sci 20 (2013), 6.

45.

Forghanifard

M.M.

Ardalan Khales

Javdani-Mallak

Rad

Farshchian

and Abbaszadegan

M.R.

, Stemness state regulators SALL4 and SOX2 are involved in progression and invasiveness of esophageal squamous cell carcinoma, Med Oncol 31 (2014), 922.

First	Detecting	Specimen	Cut-off	Sample size		HR (95% CI)
author	method		value	Higher	Lower	OS	DFS/RFS/PFS
				expression	expression
He	IHC	Tumor	Score $\geqslant$ 5	28	62	2.13 (0.592–7.667)M	3.312 (1.85–7.146)M
Yong1	TMA	Tumor	Median	114	114	1.52 (1.00–2.32)M	1.67 (1.11–2.51)M
Yong2	IHC	Tumor	Score $\geqslant$ 1	45	34	2.87 (1.09–7.52)M	2.7 (0.41–18.72)M
Oikawa	TMA	Tumor	Median	55	55	1.51 (0.75–3.04)U*	NA
Zeng	IHC	Tumor	$>$ 5% cells stained	101	43	NA	2.74 (1.10–6.80)U*
Shibahara	IHC	Tumor	$>$ 2% cells stained	47	290	1.155 (0.591–2.555)U	0.843 (0.457–1.554)M
Liu	IHC	Tumor	Score $\geqslant$ 3	59	25	3.07 (1.29–7.28)U*	NA
Han	ELISA	Serum	$\geqslant$ 130.43 pg/mL	29	27	2.16 (1.03–4.49)U*	NA
Park	IHC	Tumor	$>$ 10% cells stained	39	151	3.556 (1.511–8.367)M	NA
Yue	IHC	Tumor	Score $>$ 5	103	57	1.85 (0.74–4.62)U*	NA
Lei	IHC	Tumor	Score $\geqslant$ 4	56	24	1.16 (0.34–3.93)U*	NA
Tanaka	IHC	Tumor	$\geqslant 5\%$ cells stained	8	82	2.15 (0.94–4.95)U*	1.05 (0.46– 2.43)U*
Zhang	IHC	Tumor	Score $\geqslant$ 2	39	15	6.416 (2.053–20.052)M	NA
Kilic1	IHC	Tumor	Score $\geqslant$ weak	14	138	2.38 (1.48–3.82)U*	NA
Kilic2	IHC	Tumor	Score $\geqslant$ weak	22	93	NA	1.19 (0.68–2.09)U*
Yang	IHC	Tumor	Score $\geqslant$ 2	53	38	2.188 (1.242–4.379)M	2.961 (1.614–4.721)M

Higher expression of SALL4 predicts poor cancer prognosis: A meta-analysis

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Literature search

2.2 Inclusion criteria

2.4 Study quality assessment

2.5 Statistical analysis

3.1 Characteristics of the studies

3.2 Evidence synthesis

3.2.1 SALL4 expression and OS

3.3 Sensitivity analysis

3.4 Publication bias

Abbreviations

Footnotes

Acknowledgments

Conflict of interest

References