Lgr5 High /DCLK1 High phenotype is more common in early stage and intestinal subtypes of gastric carcinomas

Abstract

BACKGROUND:

Gastric carcinoma is the third most common malignancy and is one of the main causes of cancer deaths worldwide. Cancer stem cells (CSCs) are a subpopulation of tumour cells capable of self-renewal and differentiation, likely responsible for the initiation, recurrence, metastasis and chemo/radio-resistance.

OBJECTIVE:

This study was conducted to evaluate the expression patterns and clinicopathologic significance of putative CSC markers, Lgr5 and DCLK1, in gastric carcinoma.

METHOD:

The expression levels of Lgr5 and DCLK1 were examined in a well-defined series of gastric carcinoma tissues, including 75 (80%) from intestinal and 19 (20%) from diffuse subtypes, using tissue microarray (TMA). In addition, the correlation of the expression of these markers with clinicopathological factors was explored.

RESULTS:

Higher expressions of Lgr5 and DCLK1 were mainly detected in intestinal subtypes of gastric carcinomas compared to diffuse subtypes ( $P=$ 0.005 and $P=$ 0.050, respectively). We also found a higher expression of Lgr5 and DCLK1 more frequently in well-differentiated gastric carcinoma cases ( $P<$ 0.001 and $P=$ 0.007). The combined analysis demonstrated that the co-expression of Lgr5 and DCLK1 (Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ ) was more common in intestinal subtypes ( $P=$ 0.025) and well-differentiated gastric carcinoma samples ( $P<$ 0.001). Interestingly, there was a significant correlation between Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ phenotype and early-stage gastric carcinoma specimens ( $P=$ 0.045).

CONCLUSION:

Our findings indicated that the Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ expression pattern may be considered as a signature phenotype for intestinal subtypes of gastric carcinoma.

Keywords

Gastric carcinoma cancer stem cells Lgr5 DCLK1

1. Introduction

Gastric carcinoma is the third most common malignancy and is considered as one of the main causes of cancer deaths around the world [1, 2]. Adenocarcinomas comprise 90% of all gastric carcinoma cases and can be broadly categorized into two distinct histological subtypes: intestinal and diffuse, with the latter form exhibiting a specific molecular profile and highly aggressive behavior [2]. With a global 5-year survival rate of 30%, gastric carcinoma has a low prognosis level, with most patient deaths being from metastasis and treatment failure [3, 4]. Generally, diagnosis of gastric carcinomas happens at an advanced stage, leaving little room for effective clinical intervention, including targeted therapy; thus urgent attention to improve that available treatment method is required [3]. It is increasingly difficult to overlook the role of cancer stem cells (CSCs) in the initiation, metastasis, recurrence, and chemo-resistance of tumours, giving them considerable clinical significance, largely credited to their role in tumour angiogenesis [3, 5, 6]. Various potential stem cell markers including CD44, CD24, C133 and ALDH1 have been used for identifying CSCs in gastric carcinoma, among them leucine-rich repeat-containing G-protein coupled receptor (Lgr5) has been proposed as the most potent gastrointestinal CSC marker [7, 8, 9, 10, 11, 12, 13, 14]. Some previous studies indicated a positive correlation between the higher expression of Lgr5, as a target of the Wnt signaling pathway, with lymph node metastasis, distant seeding and shorter survival in gastric carcinoma, whereas other reports showed increase expression of Lgr5 in the intestinal subtype, well-differentiated and early stage gastric carcinomas [8, 9, 12, 13, 14, 15, 16]. Ablation of Lgr5 expression in gastric carcinoma cells demonstrated a reduction of cell invasion and increased sensitivity to chemotherapeutic agents, while it has been shown that the overexpression of Lgr5 had no significant effect on cell growth and migration [10, 13, 14].

Doublecortin-like kinase 1 (DCLK1) is a microtu-bule-associated protein that has received much attention as a potential CSC marker that distinguishes them from normal stem cells in colorectal carcinoma [11, 17, 18, 19, 20]. Nakanishi et al showed that DCLK1+ colorectal cancer cells mark a subset of tumour cells with the higher potential for tumour initiation, sphere formation assay and in vivo tumorigenicity [11]. DCLK1 can also accelerate intestinal tumour growth and tumourigenesis in mice with an Apc mutation (Apc ${}^{\text{Min /+}}$ ) [21]. Pancreatic tumour cells expressing DCLK1 in nude mice exhibited spheroid formation and tumorogenesis [17, 18]. Targeting of DCLK1-positive tumour cells in pancreatic, colorectal, and hepatocellular carcinoma showed that this molecule may be a reliable candidate molecule in the targeted therapy of these malignancies [11, 22, 23]. Therapeutic approaches based on cancer stem cell hypothesis have focused on identifying specific CSC markers to design novel agents for selective targeting of these marker-bearing CSCs [3, 16]. Considering the lack of consensus between previous studies regarding the potency of Lgr5 as a CSC marker in gastric carcinoma, its expression, and clinical significance was examined in a comprehensive collection of gastric carcinoma samples. In addition, we have evaluated the potential CSC marker DCLK1 in the same series of gastric carcinoma samples given the considerable similarities that exist between colorectal and gastric tumours and the fact that DCLK1 has been proposed as a distinguishable colorectal CSC marker. In addition, while DCLK1 can be applied as a valuable molecule in tailored therapy, while there is no available data regarding DCLK1 expression and its clinical significance in gastric carcinoma [7, 11, 24]. While a strong association has been reported between the Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ phenotype with CSC features in colorectal carcinomas; this study has investigated, for the first time, the correlations between the expressions of these CSC markers with clinicopathological parameters in gastric carcinoma samples [7, 11, 25].

2. Materials and methods

2.1 Study population

The current study was carried out on documents of gastric carcinoma patients who were admitted to two major referral hospitals in Tehran, Iran in two consecutive years (2008 and 2009). This study has included clinicopathological data from 94 gastric carcinoma patients that had not received any additional preoperative treatment. The clinicopathological parameters, including age, gender, histological subtype, tumour stage, tumour differentiation, depth of invasion, regional lymph node and omental involvement, tumour location, vascular and neural invasion were documented in databases and kept fully anonymous throughout the course of the study (Table 1). The TNM stage in gastric carcinoma is a valid prognostic factor that is comprised of five groups; stage 0, I and II are associated with five-year survival rates more than 90%, 50%, and 30%, respectively, whereas five-year survival rates in stage III and IV are less than 20% [26]. In this study, we classified gastric carcinomas based on the Lauren classification (1965) as well as the WHO classification (2010). The Lauren classification includes intestinal, diffuse and intermediate types, whereas the WHO classification includes tubular, papillary, mucinous and poorly differentiated adenocarcinomas [27].

Table 1
Correlation between Lgr5 and DCLK1 expressions and clinicopathological features in gastric cancer patients ( $P$ -value; Pearson $\chi$ 2). The values shown in bold, are statistically significant

Clinicopathologic	Total	Lgr5 expression		$P$ -value	DCLK1 expression		$P$ -value
characteristics	number	(Mean $H$ -score $=$ 106)			(Mean $H$ -score $=$ 218)
	(%)	Low	High		Low	High
Mean age (ranged: 30–100 years)
$\leqslant$ 63	44(47)	27(61)	17(39)	0.039	20(45)	24(55)	0.594
$>$ 63	50(53)	20(40)	30(60)		20(40)	30(60)
Gender
Male	77(82)	36(47)	41(53)	0.180	33(43)	44(57)	0.899
Female	17(18)	11(65)	6(35)		7(41)	10(59)
Depth of invasion
pT1–T2	42(45)	20(48)	22(52)	0.678	20(48)	22(52)	0.372
pT3–T4	52(55)	27(52)	25(48)		20(38)	32(62)
Lymph node metastasis ${}^{\rm a}$
N 0–1	64(69)	30(47)	34(53)	0.294	27(42)	37(58)	0.812
N 2–3	29(31)	17(59)	12(41)		13(45)	16(55)
Tumor stage (TNM)
I	15(16)	5(33)	10(67)	0.194	7(47)	8(53)	0.159
II	22(23)	10(45)	12(55)		8(36)	14(64)
III	48(51)	25(52)	23(48)		24(50)	24(50)
IV	9(10)	7(78)	2(22)		1(11)	8(89)
Lauren classification
Intestinal	75(80)	32(43)	43(57)	0.005	29(39)	46(61)	0.050
Diffuse	19(20)	15(79)	4(21)		11(58)	8(42)
Histological subtypes
Tubular adenocarcinoma	10(11)	3(30)	7(70)	0.020	0(0)	10(100)	0.011
Papillary adenocarcinoma	58(62)	27(47)	31(53)		25(43)	33(57)
Mucinous adenocarcinoma	7(7)	2(29)	5(71)		4(57)	3(43)
Poorly differentiated adenocarcinoma	19(20)	15(79)	4(21)		11(58)	8(42)
Tumour differentiation
Well	10(11)	3(30)	7(70)	$<$ 0.001	0(0)	10(100)	0.007
Moderate	33(35)	8(24)	25(76)		13(39)	20(61)
Poor	51(54)	36(71)	15(29)		27(53)	24(47)
Tumour location
Cardia & body	52(55)	26(50)	26(50)	1	18(35)	34(65)	0.083
Antrum & diffuse	42(45)	21(50)	21(50)		22(52)	20(48)
Vascular invasion
Yes	53(56)	29(55)	24(45)	0.298	23(43)	30(57)	0.851
No	41(44)	18(44)	23(56)		17(41)	24(59)
Neural invasion ${}^{\rm a}$
Yes	55(59)	31(56)	24(44)	0.164	22(40)	33(60)	0.356
No	34(36)	14(41)	20(59)		17(50)	17(50)
Omental involvement ${}^{\rm a}$
Yes	6(9)	5(83)	1(17)	0.094	3(50)	3(50)	0.612
No	61(91)	29(47)	32(53)		24(39)	37(61)

${}^{\rm a}$ : The analysis of these characteristics was conducted with available patient data.

2.2 Tissue Microarray (TMA) construction

The representative area of each tumour tissue was identified and marked out by a pathologist from Hematoxylin and Eosin (H&E) stained slides as described previously [28, 29]. The marked tissue samples were then arrayed onto a recipient block with a diameter of 0.6 mm by using a Tissue Arrayer Minicore (ALPHELYS, Plaisir, France). Finally, the tissue microarray blocks were cut into 4 $\mu$ m thin sections and placed on positively charged TMA slides (Superfrost plus, Thermo Scientific, Germany). TMA blocks were constructed in three copies for each specimen; the mean scoring of three cores was then calculated as the final score. A validation study of TMA showed that despite the variability of antigen expression between cores, analysis of each core would represent more than 90% of staining pattern of the whole tissue section, whereas analysis of the two readable cores achieved greater than 95% accuracy [30]. In the present study, three cores were evaluated from each tumour to increase the accuracy and validity of the analysis and to overcome the heterogeneity in antigen expression.

2.3 Immunohistochemistry

The formalin-fixed, paraffin-embedded sections of the TMA constructed slides were stained by a standard chain polymer-conjugated technique, Envision, as described previously [31, 32, 33]. The tissue sections were dewaxed, incubated in 3% hydrogen peroxidase (20 minutes), and antigen retrieval was carried out in an autoclave for 10 minutes. The slides were then incubated with primary antibodies; rabbit polyclonal anti-Lgr5 (1:1000 dilution, ab71225; Abcam, UK) and rabbit polyclonal anti-DCLK1 (1:800 dilution, ab31704; Abcam, UK) overnight at 4 ${}^{\circ}$ C. The sections were stained and visualized using a standard EnVision-HRP kit (DAKO, Denmark). Human colorectal cancer tissues were used as a positive control for both primary antibodies and the exclusion of primary antibody and its replacement with TBS was applied as the negative control [34].

2.4 Scoring system

To evaluate Lgr5 and DCLK1 expressions, a pathologist was consulted for the diagnosis of the samples and scoring of immunostained slides was performed semi-quantitatively by two observers after a series were examined on a double-headed microscope. The obtained results were also confirmed by a pathologist (A K) to receive a comprehensive concept of staining in tumour cells as described previously [35]. This is a blinded study, therefore both the pathologist and the observer were unaware of the outcomes of the samples.

The staining of Lgr5 and DCLK1 was scored using a semi-quantitative scoring system in a coded manner, blinded to clinical and pathological parameters [29, 36, 37]. For this purpose, the immunostainig of Lgr5 and DCLK1 was evaluated by two expert investigators (ZM and AK) on a double-headed microscope and a consensus agreement was achieved. Initially, TMA slides were scanned at 10 $\times$ magnification to obtain a general impression of the overall distribution of the tumour cells. The positive cells were then assessed semi-quantitatively at higher magnifications and final scores were given respectively. The level of expression of Lgr5 and DCLK1 in gastric carcinoma was assessed by three scoring methods, namely the intensity of the staining, the percentage of positive cells and the $H$ -score. The intensity of the immunostaining was classified into four categories: 0 (no immunostaining present), 1 (weak immunostaining), 2 (moderate immunostaining) and 3 (strong immunostaining). The percentage of positive cells was assessed semi quantitatively and scored as 0 to 100%. The Histochemical score ( $H$ -score) of immunoreactivity was obtained by multiplying the intensity and percentage scores (0–100%), and a final score of 0–300 was given [38]. The mean $H$ -score was selected as a cut-off point (for Lgr5 $H$ -score $=$ 106 and DCLK1 $H$ -score $=$ 218) and specimens were categorized into two groups, higher and lower expression, in comparison to the cut-off point, as used in previous studies [39, 40].

2.5 Statistical analysis

The findings were analyzed using SPSS software version 16 (SPSS, Chicago, IL, USA). Pearson’s Chi-square and Spearman’s correlation tests were applied to evaluate the correlation between Lgr5 and DCLK1 expressions and clinicopathological parameters. The Mann-Whitney U-test was used to compare the expression of Lgr5 and DCLK1 in three tumour-differentiation groups. A difference of $P<$ 0.05 between groups was considered significant.

3. Results

3.1 Clinicopathological information of patients

Of 94 gastric carcinoma samples examined in the current study, 75 cases (80%) were intestinal subtype and 19 (20%) were diffuse subtype. The majority of patients were male, 77 (82%), and ages ranged between 30–100 years (mean $=$ 63 years). In terms of tumour differentiation, out of 94 overall cases, 10 (11%), 33 (35%), and 51 (54%) cases had well, moderate, and poor differentiation, respectively. Based on the TNM staging system, 15 (16%) of gastric carcinoma cases placed in stage I, 22 (23%) in stage II, 48 (51%) in stage III, and 9 (10%) in stage IV. Out of 93 patients with available regional lymph node involvement data, 64 (69%) showed N0–N1, and 29 (31%) had N2–N3 categories. Fifty-three (56%) gastric carcinoma cases showed vascular invasion, whereas 41 (44%) cases had no vascular invasion. All the recorded patient clinicopathological data are summarized in Table 1.

Figure 1.

Expression of Lgr5 and DCLK1 in gastric carcinoma samples. Expression levels of Lgr5 and DCLK1 showed weak, moderate, and strong intensity of staining (All images were taken at 200x magnification).

Figure 2.

Box-plot diagram of Lgr5 and DCLK1 expressions in histological subtypes and tumour differentiation of gastric carcinoma specimens. Each box-plot shows median (bold line), interquartile line (box), and outliers (circle).

3.2 Lgr5 expression and its association with clinicopathological parameters in gastric carcinoma

Immunodetection of Lgr5 expression was generally localized in the cytoplasmic or membranous area of tumour cells. In terms of $H$ -score(mean $=$ 106), 50% of cases (47/94) displayed lower Lgr5 expression ( $H$ -score $\leqslant$ 106) and 50% showed higher expression of Lgr5 (Fig. 1). Fourty three (57%) out of 75 patients with intestinal subtype of gastric carcinoma, displayed higher Lgr5 expression levels, whereas 4 (21%) out of 19 patients with the diffuse subtype of gastric carcinoma cases had a higher Lgr5 expression level. The obtained results demonstrate a higher expression of Lgr5 in intestinal subtype tumours compared to diffuse subtypes (Lauren classification, $P=$ 0.005) (Fig. 2). It was also shown that tubular adenocarcinoma displayed an increased Lgr5 expression compared to poorly differentiated adenocarcinoma (WHO classification, $P=$ 0.020). Among 10 well-differentiated gastric carcinoma specimens, 70% exhibited a higher Lgr5 expression level; and among 51 poorly differentiated gastric carcinoma samples, 29% expressed higher levels of Lgr5. Our analysis indicated that well-differentiated gastric carcinoma cases have higher levels of Lgr5 expression ( $P<$ 0.001) (Table 1). The Mann-Whitney $U$ -test was used to compare the differences between expressions of Lgr5 in three tumour differentiation groups, which showed a statistically significant difference between the expression of Lgr5 in poorly differentiated cases relative to well ( $P=$ 0.015) and moderately ( $P<$ 0.001) differentiated tumours (Fig. 2). There was no significant correlation between Lgr5 expression and other clinical data, which have been summarized in Table 1.

3.3 DCLK1 expression and its association with clinicopathological parameters in gastric carcinoma

Immunohistochemical findings showed the main localization of DCLK1 expression to be in the cytoplasmic area of gastric tumour cells. Out of 94 overall cases, a lower expression of DCLK1 (mean of $H$ -score $=$ 218) was found in 40 (43%) cases, while higher expression of DCLK1 was seen in 54 (57%) specimens (Fig. 1). The relation between DCLK1 expression and clinicopathologic parameters in gastric carcinoma was explored using Chi-square test. Of the 75 intestinal and 19 diffuse gastric carcinoma subtypes, 61% and 42% showed higher expression level of DCLK1, respectively (Fig. 2). There was a significant correlation between DCLK1 expression and tumour subtypes, indicating a higher expression of DCLK1 in intestinal subtypes (Lauren classification, $P=$ 0.050). The tubular adenocarcinoma displayed elevated levels of DCLK1 expression compared to poorly differentiated adenocarcinoma (based on the WHO classification, $P=$ 0.011). All 10/10 (100%) well-differentiated, 20/33 (61%) moderately-differentiated, and 24/51 (47%) poorly-differentiated gastric carcinoma cases showed higher DCLK1 expression levels. An inverse significant correlation was observed between DCLK1 expression and tumour-differentiation, signifying a higher DCLK1 expression in well differentiated gastric carcinomas ( $P=$ 0.007). The Mann-Whitney U test displayed a significant difference between the expression of DCLK1 in well and poorly-differentiated gastric carcinoma groups ( $P=$ 0.002) (Fig. 2).

Figure 3.

The correlation between immunohistochemical expressions of Lgr5 and DCLK1 in gastric carcinoma specimens. Expression of Lgr5 with DCLK1 showed a positive significant correlation ( $r=$ 0.120, $P=$ 0.010).

Table 2

Correlation between Lgr5/DCLK1 phenotypes and clinicopathological features in gastric cancer patients. The values shown in bold, are statistically significant

Clinicopathologic	Total	Lgr5/DCLK1 phenotypes				$P$ -value
characteristics	Number	Lgr5 ${}^{\text{Low}}$ /	Lgr5 ${}^{\text{High}}$ /	Lgr5 ${}^{\text{Low}}$ /	Lgr5 ${}^{\text{High}}$ /
	(%)	DCLK1 ${}^{\text{Low}}$	DCLK1 ${}^{\text{Low}}$	DCLK1 ${}^{\text{High}}$	DCLK1 ${}^{\text{High}}$
Mean age (ranged: 30–100 years)
$\leqslant$ 63	42(47)	15(34)	5(11)	12(27)	12(27)	0.239
$>$ 63	50(53)	11(22)	9(18)	9(18)	21(42)
Gender
Male	77(82)	22(29)	11(14)	14(18)	30(39)	0.152
Female	17(18)	4(23)	3(18)	7(41)	3(18)
Depth of invasion
pT1–T2	42(45)	13(31)	7(17)	7(17)	15(35)	0.679
pT3–T4	52(55)	13(25)	7(13)	14(27)	18(35)
Lymph node metastasis ${}^{\rm a}$
N 0–1	64(69)	17(27)	10(16)	13(20)	24(38)	0.756
N 2–3	29(31)	9(31)	4(13)	8(28)	8(28)
Tumour stage (TNM)
I	15(16)	5(33)	2(13)	0(0)	8(54)	0.045
II	22(23)	4(18)	4(18)	6(27)	8(37)
III	48(51)	16(33)	8(17)	9(19)	15(31)
IV	9(10)	1(11)	0(0)	6(67)	2(22)
Lauren classification
Intestinal	75(80)	15(20)	14(19)	17(23)	29(39)	0.004
Diffuse	19(20)	11(58)	0(0)	4(21)	4(21)
Histological subtypes
Tubular adenocarcinoma	10(11)	0(0)	0(0)	3(30)	7(70)	0.006
Papillary adenocarcinoma	58(62)	13(22)	12(21)	14(24)	19(33)
Mucinous adenocarcinoma	7(7)	2(29)	2(29)	0(0)	3(42)
Poorly differentiated adenocarcinoma	19(20)	11(58)	0(0)	4(21)	4(21)
Tumour differentiation
Well	10(11)	0(0)	(0)	3(30)	7(70)	$<$ 0.001
Moderate	33(35)	4(12)	9(27)	4(12)	16(49)
Poor	51(54)	22(43)	5(10)	14(27)	10(20)
Tumour location
Cardia & body	52(55)	13(25)	5(10)	13(25)	21(40)	0.296
Antrum & diffuse	42(45)	13(31)	9(21)	8(19)	12(29)
Vascular invasion
Yes	53(56)	16(30)	7(13)	13(25)	17(32)	0.788
No	41(44)	10(24)	7(17)	8(19)	16(39)
Neural invasion ${}^{\rm a}$
Yes	55(62)	16(29)	6(11)	15(27)	18(33)	0.305
No	34(38)	9(27)	8(23)	5(15)	12(35)
Omental involvement ${}^{\rm a}$
Yes	6(9)	3(50)	0(0)	2(33)	1(17)	0.409
No	61(91)	16(26)	8(13)	13(21)	24(39)

${}^{\rm a}$ : The analysis of these characteristics was conducted with available patient data.

3.4 Combined analysis of Lgr5 and DCLK1 expressions in gastric carcinoma

Comparing the expression patterns of Lgr5 and DCLK1 markers demonstrated a reciprocal significant correlation between the two markers in the same series of gastric carcinoma samples ( $P=$ 0.010, Fig. 3). Among 94 gastric carcinoma samples, 26 (28%) cases showed the Lgr5 ${}^{\text{low}}$ /DCLK1 ${}^{\text{low}}$ phenotype, 14 (15%) cases had the Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{low}}$ phenotype, 21 (22%) showed the Lgr5 ${}^{\text{low}}$ /DCLK1 ${}^{\text{high}}$ phenotype and 33 (35%) cases had the Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ phenotype (Table 2). The one-way ANOVA and Tukey’s post hoc analysis was used to examine the correlation between expressions of Lgr5/DCLK1 phenotypes with clinicopathological parameters. The analysis showed a significant correlation between Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ phenotype and intestinal subtype of gastric carcinoma (Lauren classification, $P=$ 0.004). The Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ phenotype was more often found in tubular adenocarcinoma compared to poorly differentiated adenocarcinoma (WHO classification, $P=$ 0.006). A highly significant correlation was also seen between the Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ phenotype and well-differenti-ated gastric carcinoma cases ( $P<$ 0.001). Similarly, co-expression of Lgr5 and DCLK1 (Lgr5 ${}^{\text{high}}$ /DCLK1 ${}^{\text{high}}$ ) was frequently detected in the early stage of gastric carcinoma ( $P=$ 0.045). There was no significant association between the expression of Lgr5/DCLK1 phenotypes and other clinicopathological variables (Table 2).

4. Discussion

A number of recent studies have indicated the critical role of cancer stem cells (CSCs), in the initiation, maintenance, drug resistance, cell invasion, and metastasis of gastric carcinoma [6]. Identification of definite gastric CSC markers would significantly enhance the current understanding of CSC biology. This would pave the way for overcoming multidrug resistance, which is one of the greatest obstacles in the effective treatment of gastric carcinoma.

Lgr5, a G-protein coupled receptor, is a member of the Wnt signaling pathway with an unknown ligand, and DCLK1 is a microtubule-associated protein, both of which have been proposed as the most putative gastrointestinal CSCs markers [11, 18]. In the present study, the expressions of putative CSC markers, Lgr5 and DCLK1, were examined in gastric carcinoma cases. Our immunohistochemical analysis demonstrated a higher expression of Lgr5 in the intestinal subtype compared to diffuse subtype of gastric carcinoma. Increased expression of Lgr5 was mainly detected in well-differentiated tumours as opposed to poorly differentiated tumours. Further supporting our results, Bu et al. showed elevated Lgr5 expression in intestinal subtypes of gastric carcinoma samples [8]. They also demonstrated higher Lgr5expression in well-differentiated and early-stage gastric carcinoma tissues [8]. In a previous study, both gene and protein expression of Lgr5 were evaluated and reported in well differentiated gastric tumour cells, while Lgr5 expression was not detected in poorly differentiated samples [12]. More recently, the real time-PCR and RNA in situ hybridization analysis of Lgr5 showed higher expression of Lgr5 in well-differentiated gastric carcinoma [10]. It has been demonstrated that Lgr5 can act as a stem cell marker in the early stage of gastric carcinoma intestinal subtypes [41]. Similarly, in colorectal carcinomas, an inverse significant association between Wnt target genes, including Lgr5, and favorable outcomes have been demonstrated [42]. In contrast, the spatial distribution of Lgr5 expression has been reported to play a key role in gastric carcinoma tumorogenesis [9]. Previous literature indicate a positive significant correlation between lymph node involvement, distant metastasis and poor survival [13, 14, 43, 44]. More recent reports have shown that gastric carcinoma cells expressing Lgr5 exhibited CSC properties, including decreased apoptosis, drug resistance and in vivo tumorogenesis [14, 45]. Mounting evidence revealed that Lgr5 is overexpressed in a very early stage in tumorigenesis and its expression declines during tumour progression [46, 47]. In a pioneer study, higher expression of Lgr5 was found in early stages (I and II) of colorectal and ovarian cancers and decreased in advanced stages (III and IV) in tissue samples [46]. The expression patterns of Lgr5 in the mouse endometrial cancer model also showed a declined trend from early to advanced stages [47]. Control of gene expression is a complex and multi-faceted process; thus the gene regulation of Lgr5 may lead to different levels of protein expression. Wnt target genes are regulated by epigenetic modifications, with the methylation of CpG islands in the promoter region being the most well-known process [42, 48]. In agreement with our findings, Sousa et al reported that hypermethylation of the Lgr5 promoter leads to decreased expression of Lgr5 during tumour progression of colorectal cancer [42]. It has been also suggested that dysregulation of Lgr5 alone is not enough for gastric tumorogenesis, and activation of the Wnt signaling pathway can provide fertile ground for the process [10]. Jang et al. (2016) showed that there is no correlation between Lgr5 expression with survival, whereas the enhanced activity of Wnt signaling pathway parallel with up-regulation of Lgr5 declined gasric cancer patient outcomes [10].

DCLK1 is a serine/threonine-protein kinase that has been suggested as the most specific colorectal CSC marker that clearly distinguishes CSCs from normal stem cells (NSCs) [11]. In this study, for the first time, we evaluated the expression of DCLK1 in gastric carcinoma cases as well as its potential clinical significance. Our findings showed similar results to Lgr5 markers, indicating increased expression of DCLK1 in the intestinal subtype and well-differentiated gastric carcinoma samples. There is an absence of robust studies on DCLK1 expression in gastric carcinoma, but there is a large body of studies about the analysis of potential CSC DCLK1 in pancreatic and colorectal cancers [7, 11], which can be used for the purpose of our comparison. In line with our findings, a more recent study has demonstrated that DCLK1 expression has a strong tendency to decrease with pancreatic tumour progression from early stages (I–II) to advanced stages (III–IV) [49]. In a study conducted by Tao et al., DCLK1 expression was found in stage I of Non-Small Cell Lung Cancer (NSCLC) and correlated with poor survival [50]. It has been shown that inhibition of DCLK1 in pancreatic and colorectal cancer cells can led to the downregulation of Epithelial-mesenchymal Transition (EMT)-related factors, including ZEB1, ZEB2, SNAIL, and SLUG [19, 22, 51, 52]. EMT is a prerequisite for tumour cells invasion and metastasis that is orchestrated by several pre-existing factors [53]. During EMT, epithelial factor expression diminishes and consequently, an increase in mesenchymal factors is observed [53]. In our study, the expression of EMT-related markers was not analyzed in the gastric carcinoma tissues, but it can be a consideration for future investigation.

DCLK1 is a differentiation marker and decreased expression of DCLK1 is a probable alternation in the course of development and progression of aggressive behavior in tumours [49, 54]. This leads to an understanding that DCLK1 is a worthwhile focus for studies on gastric carcinoma tumorigenesis and metastasis. Recently, hypermethylation of the 5 ${}^{\prime}$ promoter region of the DCLK1 gene was reported in some studies, emphasizing that the silencing of DCLK1 is a very early step during adenoma-carcinoma progression of human colon carcinogenesis [55, 56, 57]. In addition, two isoforms have been reported for DCLK1, including long ( $\sim$ 80–82 KDa) and short ( $\sim$ 45–50 KDa) isoforms. The long isoform consists of two N-terminals, one C- terminal and one middle domain (serine/proline rich) which mediates protein interactions, whereas the short isoform lacks one of the two N-terminal domains [56]. The available commercial antibodies were designed against the C-terminal domain; thus, the identification of the specific isoform(s) expression patterns in cancer tissues, such as gastric carcinomas, remains largely unknown. Therefore, the application of DCLK1 for tailored therapy of gastrointestinal cancers warrants further evaluation.

To identify a more representative CSC population in gastric carcinoma, we used the combined analysis of both Lgr5 and DCLK1 markers. The co-expression of Lgr5 and DCLK1 expressions (Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ ) were frequently observed in intestinal subtype, well-differentiated and early stage gastric carcinoma. Intestinal subtypes are the most common forms of early and well-differentiated gastric carcinomas; therefore, Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ patterns may be a signature recognition pattern for these subtypes [27]. A major difference between DCLK1 and Lgr5 is that, under basal conditions, the former is rapidly expressed in cycling mitotic cells, while the latter is restricted to quiescent cells, and only expressed in mitotic cells after irradiation [52]. Hence, studying DCLK1 and Lgr5 expressing cells would simultaneously allow the examination of the gene expression profiles and molecular signatures, which in turn could lead to the development of novel regenerative strategies for gastric carcinoma treatment [52].

The study of CSC markers is still a novel avenue of cancer research, especially for gastric carcinoma, and our study has evaluated the co-expression of Lgr5 and DCLK1 markers on gastric carcinoma specimens for the first time. Preceding similar studies on Lgr5 and DCLK1 have not investigated their expression in conjunction with each other or alongside known biomarkers [7]. Thus, for future studies, it is suggested that the distribution of other potential CSC markers, including CD44, CD24, CD133, ALDH1, be analyzed in combination with Lgr5 and DCLK1 markers on gastric carcinoma tissues to evaluate a more comprehensive prognostic and therapeutic strategy.

5. Conclusions

Our findings denote that the co-expression of potential CSC markers Lgr5 and DCLK1 or Lgr5 ${}^{\text{High}}$ /DCLK1 ${}^{\text{High}}$ expression pattern may be considered as a signature phenotype for early stage, well differentiation, and intestinal subtypes of gastric carcinoma. Further analysis is required for the investigation of CSC markers Lgr5 and DCLK1 as a potential therapeutic target for gastric carcinoma.

Conflict of interest

The authors declare that they have no conflict of interest.

Footnotes

Acknowledgments

This study was supported by a grant from Iran University of Medical Sciences (Grant #12417-12-04-89).

References

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer Statistics, CA Cancer J Clin 66(1) (2016), 7–30.

McLean

M.H.

and El-Omar

E.M.

, Genetics of gastric cancer, Nat Rev Gastroenterol Hepatol 11(11) (2014), 664–674.

Zhang

Hua

Wang

Huang

Gan

Zhang

Wang

Cheng

and Guo

, Identification of stem-like cells and clinical significance of candidate stem cell markers in gastric cancer, Oncotarget 7(9) (2016), 9815–9831.

Thun

M.J.

DeLancey

J.O.

Center

M.M.

Jemal

and Ward

E.M.

, The global burden of cancer: priorities for prevention, Carcinogenesis 31(1) (2010), 100–110.

Dou

and Gu

, Emerging strategies for the identification and targeting of cancer stem cells, Tumour Biol 31(4) (2010), 243–253.

Tirino

Desiderio

Paino

De Rosa

Papaccio

La Noce

Laino

De Francesco

and Papaccio

, Cancer stem cells in solid tumors: an overview and new approaches for their isolation and characterization, FASEB J 27(1) (2013), 13–24.

Brungs

Aghmesheh

Vine

K.L.

Becker

T.M.

Carolan

M.G.

and Ranson

, Gastric cancer stem cells: evidence, potential markers, and clinical implications, J Gastroenterol 51(4) (2016), 313–326.

Zheng

Zhang

Sun

Dong

Wang

Cheng

Xing

and Ji

, LGR5 is a promising biomarker for patients with stage I and II gastric cancer, Chin J Cancer Res 25(1) (2013), 79–89.

Simon

Petke

Böger

Behrens

H.M.

Warneke

Ebert

and Röcken

, The spatial distribution of LGR5+ cells correlates with gastric cancer progression, PLoS One 7(4) (2012), e35486.

10.

Jang

B.G.

Lee

B.L.

and Kim

W.H.

, Prognostic significance of leucine-rich-repeat-containing G-protein-coupled receptor 5, an intestinal stem cell marker, in gastric carcinomas, Gastric Cancer 19(3) (2016), 767–777.

11.

Nakanishi

Seno

Fukuoka

Ueo

Yamaga

Maruno

Nakanishi

Kanda

Komekado

Kawada

Isomura

Kawada

Sakai

Yanagita

Kageyama

Kawaguchi

Taketo

M.M.

Yonehara

and Chiba

, Dclk1 distinguishes between tumor and normal stem cells in the intestine, Nat Genet 45(1) (2013), 98–103.

12.

Xie

Gao

Wang

Guo

Tian

and Fan

, Lgr5 expression as stem cell marker in human gastric gland and its relatedness with other putative cancer stem cell markers, Gene 525(1) (2013), 18–25.

13.

H.Q.

Cai

A.Z.

X.S.

Cui

J.X.

Shen

W.S.

Bian

S.B.

Wang

J.Y.

C.R.

Song

Wei

and Chen

, Leucine-rich repeat-containing G-protein-coupled receptor 5 is associated with invasion, metastasis, and could be a potential therapeutic target in human gastric cancer, Br J Cancer 110(8) (2014), 2011–2020.

14.

H.Q.

Cui

J.X.

Shen

W.S.

X.S.

Bian

S.B.

J.Y.

Song

Wei

and Chen

, Increased expression of Lgr5 is associated with chemotherapy resistance in human gastric cancer, Oncol Rep 32(1) (2014), 181–188.

15.

Van der Flier

L.G.

Sabates-Bellver

Oving

Haegebarth

De Palo

Anti

Van Gijn

M.E.

Suijkerbuijk

Van de Wetering

Marra

and Clevers

, The intestinal Wnt/TCF signature, Gastroenterology 132(2) (2007), 628–632.

16.

Becker

Huang

and Mashimo

, Immunostaining of Lgr5, an intestinal stem cell marker, in normal and premalignant human gastrointestinal tissue, Scientific World Journal 8 (2008), 1168–1176.

17.

May

Riehl

T.E.

Hunt

Sureban

S.M.

Anant

and Houchen

C.W.

, Identification of a Novel Putative Gastrointestinal Stem Cell and Adenoma Stem Cell Marker, Doublecortin and CaM Kinase-Like-1, Following Radiation Injury and in Adenomatous Polyposis Coli/Multiple Intestinal Neoplasia Mice, Stem Cells 26(3) (2008), 630–637.

18.

May

Sureban

S.M.

Lightfoot

S.A.

Hoskins

A.B.

Brackett

D.J.

Postier

R.G.

Ramanujam

Rao

C.V.

Wyche

Anant

and Houchen

C.W.

, Identification of a novel putative pancreatic stem/progenitor cell marker DCAMKL-1 in normal mouse pancreas, Am J Physiol Gastrointest Liver Physiol 299(2) (2010), G303–G310.

19.

Sureban

S.M.

May

Lightfoot

S.A.

Hoskins

A.B.

Lerner

Brackett

D.J.

Postier

R.G.

Ramanujam

Mohammed

Rao

C.V.

Wyche

J.H.

Anant

and Houchen

C.W.

, DCAMKL-1 Regulates Epithelial – Mesenchymal Transition in Human Pancreatic Cells through a miR-200a – Dependent Mechanism, Cancer Res 71(6) (2011), 2328–2338.

20.

Vega

K.J.

May

Sureban

S.M.

Lightfoot

S.A.

Reed

Weygant

Ramanujam

Souza

Madhoun

Whorton

Anant

Meltzer

S.J.

and Houchen

C.W.

, Identification of the putative intestinal stem cell marker doublecortin and CaM kinase-like-1 in Barrett’s esophagus and esophageal adenocarcinoma, J Gastroenterol Hepatol 27(4) (2012), 773–780.

21.

Chandrakesan

Weygant

May

Chinthalapally

H.R.

Sureban

S.M.

Ali

Lightfoot

S.A.

Umar

and Houchen

C.W.

, DCLK1 facilitates intestinal tumor growth via enhancing pluripotency and epithelial mesenchymal transition, Oncotarget 5(19) (2014), 9269–9280.

22.

Sureban

S.M.

May

Weygant

Chandrakesan

Bannerman-Menson

Ali

Pantazis

Westphalen

C.B.

Wang

T.C.

and Houchen

C.W.

, XMD8-92 inhibits pancreatic tumor xenograft growth via a DCLK1-dependent mechanism, Cancer Lett 351(1) (2014), 151–161.

23.

Wang

Zhang

Wang

Zhang

and Tang

, miR-613 inhibits the growth and invasiveness of human hepatocellular carcinoma via targeting DCLK1, Biochem Biophys Res Commun 473(4) (2016), 987–992.

24.

Mirzaei

Tavoosidana

Rad

A.A.

Rezaei

Tavakoli-Yaraki

Kadijani

A.A.

Khalili

and Madjd

, A new insight into cancer stem cell markers: Could local and circulating cancer stem cell markers correlate in colorectal cancer? Tumour Biol 37(2) (2016), 2405–2414.

25.

May

Sureban

S.M.

Hoang

Riehl

T.E.

Lightfoot

S.A.

Ramanujam

Wyche

J.H.

Anant

and Houchen

C.W.

, Doublecortin and CaM Kinase-like-1 and Leucine-Rich-Repeat-Containing G-Protein-Coupled Receptor Mark Quiescent and Cycling Intestinal Stem Cells, Respectively, Stem Cells 27(10) (2009), 2571–2579.

26.

Washington

, 7th edition of the AJCC cancer staging manual: stomach, Ann Surg Oncol 17(12) (2010), 3077–3079.

27.

El Hajj

Sittler

Lammert

Barnes

and Meloni-Ehrig

, Gastric cancer: Classification, histology and application of molecular pathology, J Gastrointest Oncol 3(3) (2012), 251–261.

28.

Korourian

Roudi

Shariftabrizi

Kalantari

Sotoodeh

and Madjd

, Differential role of Wnt signaling and base excision repair pathways in gastric adenocarcinoma aggressiveness, Clin Exp Med (2016); DOI: 10.1007/s10238-016-0443-0.

29.

Mohsenzadegan

Madjd

Asgari

Abolhasani

Shekarabi

Taeb

and Shariftabrizi

, Reduced expression of NGEP is associated with high-grade prostate cancers: a tissue microarray analysis, Cancer Immunol Immunother 62(10) (2013), 1609–1618.

30.

Camp

R.L.

Charette

L.A.

and Rimm

D.L.

, Validation of tissue microarray technology in breast carcinoma, Lab Invest 80(12) (2000), 1943–1949.

31.

Keymoosi

Gheytanchi

Asgari

Shariftabrizi

and Madjd

, ALDH1 in combination with CD44 as putative cancer stem cell markers are correlated with poor prognosis in urothelial carcinoma of the urinary bladder, Asian Pac J Cancer Prev 15(5) (2014), 2013–2020.

32.

Mehrazma

Madjd

Kalantari

Panahi

Hendi

and Shariftabrizi

, Expression of stem cell markers, CD133 and CD44, in pediatric solid tumors: a study using tissue microarray, Fetal Pediatr Pathol 32(3) (2013), 192–204.

33.

Roudi

Madjd

Korourian

Mehrazma

Molanae

Sabet

M.N.

and Shariftabrizi

, Clinical significance of putative cancer stem cell marker CD44 in different histological subtypes of lung cancer, Cancer Biomark 14(6) (2014), 457–67.

34.

Gao

Zhou

J.C.

Gao

S.X.

Zhu

G.C.

Zhang

X.G.

and Yang

, The role of LGR5 and ALDH1A1 in non-small cell lung cancer: Cancer progression and prognosis, Biochem Biophys Res Commun 462(2) (2015), 91–98.

35.

Nagel

Laskawi

Wahlers

and Hemmerlein

, Expression of matrix metalloproteinases MMP-2, MMP-9 and their tissue inhibitors TIMP-1,-2, and-3 in benign and malignant tumours of the salivary gland, Histopathology 44(3) (2004), 222–231.

36.

Madjd

Durrant

L.G.

Bradley

Spendlove

Ellis

I.O.

and Pinder

S.E.

, Loss of CD55 is associated with aggressive breast tumors, Clin Cancer Res 10(8) (2004), 2797–2803.

37.

Taeb

Asgari

Abolhasani

Farajollahi

M.M.

and Madjd

, Expression of prostate stem cell antigen (PSCA) in prostate cancer: A tissue microarray study of Iranian patients, Pathol Res Pract 210(1) (2014), 18–23.

38.

McCarty

K.S.

Miller

L.S.

Cox

E.B.

Konrath

and McCarty

K.S.

, Estrogen receptor analyses. Correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies, Arch Pathol Lab Med 109(8) (1985), 716–721.

39.

Erfani

Roudi

Rakhshan

Sabet

M.N.

Shariftabrizi

and Madjd

, Comparative expression analysis of putative cancer stem cell markers CD44 and ALDH1A1 in various skin cancer subtypes, Int J Biol Markers 31(1) (2016), e53–61.

40.

Roudi

Korourian

Shariftabrizi

and Madjd

, Differential Expression of Cancer Stem Cell Markers ALDH1 and CD133 in Various Lung Cancer Subtypes, Cancer Invest 33(7) (2015), 294–302.

41.

Jang

B.G.

Lee

B.L.

and Kim

W.H.

, Distribution of LGR5+ cells and associated implications during the early stage of gastric tumorigenesis, PloS One 8(12) (2013), e82390.

42.

de Sousa E. Melo

Colak

Buikhuisen

Koster

Cameron

de Jong

J.H.

Tuynman

J.B.

Prasetyanti

P.R.

Fessler

van den Bergh

S.P.

Rodermond

Dekker

van der Loos

C.M.

Pals

S.T.

van de Vijver

M.J.

Versteeg

Richel

D.J.

Vermeulen

and Medema

J.P.

, Methylation of cancer-stem-cell-associated Wnt target genes predicts poor prognosis in colorectal cancer patients, Cell Stem Cell 9(5) (2011), 476–485.

43.

Huang

Qiu

Xiao

Wang

Zhang

and Bai

, The prognostic role of Leucine-rich repeat-containing G-protein-coupled receptor 5 in gastric cancer: A systematic review with meta-analysis, Clin Res Hepatol Gastroenterol 40(2) (2016), 246–253.

44.

Yamanoi

Fukuma

Uchida

Kushima

Yamazaki

Katai

Kanai

and Sakamoto

, Overexpression of leucine-rich repeat-containing G protein-coupled receptor 5 in gastric cancer, Pathol Int 63(1) (2013), 13–19.

45.

Wang

and Liu

, Lgr5-Positive Cells are Cancer-Stem-Cell-Like Cells in Gastric Cancer, Cell Physiol Biochem 36(6) (2015), 2447–2455.

46.

McClanahan

Koseoglu

Smith

Grein

Gustafson

Black

Kirschmeier

and Samatar

A.A.

, Identification of overexpression of orphan G protein-coupled receptor GPR49 in human colon and ovarian primary tumors, Cancer Biol Ther 5(4) (2006), 419–426.

47.

Sun

Jackson

Dey

S.K.

and Daikoku

, In pursuit of leucine-rich repeat-containing G protein-coupled receptor-5 regulation and function in the uterus, Endocrinology 150(11) (2009), 5065–5073.

48.

Basu

Haase

and Ben-Ze’ev

, Wnt signaling in cancer stem cells and colon cancer metastasis, F1000Res 19 (2016), 5.

49.

Johnson

Chandrakesan

Weygant

May

Aiello

Rhim

Zhao

Zheng

Lightfoot

Pant

Irvan

Postier

Hocker

Hanas

J.S.

Ali

Sureban

S.M.

Schlosser

M.J.

Stanger

and Houchen

C.W.

, Doublecortin-like kinase 1 is elevated serologically in pancreatic ductal adenocarcinoma and widely expressed on circulating tumor cells, PloS One 10(2) (2015), e0118933.

50.

Tao

Tanaka

and Okabe

, Doublecortin and CaM kinase-like-1 expression in pathological stage I non-small cell lung cancer, J Cancer Res Clin Oncol 143(8) (2017), 1449–1459.

51.

Sureban

S.M.

May

Mondalek

F.G.

Ponnurangam

Pantazis

Anant

Ramanujam

R.P.

and Houchen

C.W.

, Nanoparticle-based delivery of siDCAMKL-1 increases microRNA-144 and inhibits colorectal cancer tumor growth via a Notch-1 dependent mechanism, J Nanobiotechnology 9(1) (2011), 40.

52.

Sureban

S.M.

May

Ramalingam

Subramaniam

Natarajan

Anant

and Houchen

C.W.

, Selective blockade of DCAMKL-1 results in tumor growth arrest by a Let-7a MicroRNA-dependent mechanism, Gastroenterology 137(2) (2009), 649–659.

53.

Diepenbruck

and Christofori

, Epithelial–mesenchymal transition (EMT) and metastasis: yes, no, maybe? Curr Opin Cell Biol 43 (2016), 7–13.

54.

Weygant

Kaddis

J.S.

Berry

W.L.

May

Chandrakesan

Bannerman-Menson

Vega

K.J.

Tomasek

J.J.

Bronze

M.S.

and Houchen

C.W.

, Survival of patients with gastrointestinal cancers can be predicted by a surrogate microRNA signature for cancer stem-like cells marked by DCLK1 kinase, Cancer Res 76(14) (2016), 4090–4099.

55.

Vedeld

H.M.

Andresen

Eilertsen

I.A.

Nesbakken

Seruca

Gladhaug

I.P.

Thiis-Evensen

Rognum

T.O.

Boberg

K.M.

and Lind

G.E.

, The novel colorectal cancer biomarkers CDO1, ZSCAN18 and ZNF331 are frequently methylated across gastrointestinal cancers, Int J Cancer 136(4) (2015), 844–853.

56.

O’Connell

M.R.

Sarkar

Luthra

G.K.

Okugawa

Toiyama

Gajjar

A.H.

Qiu

Goel

and Singh

, Epigenetic changes and alternate promoter usage by human colon cancers for expressing DCLK1-isoforms: Clinical Implications, Sci Rep 5 (2015), 14983.

57.

Vedeld

H.M.

Skotheim

R.I.

Lothe

R.A.

and Lind

G.E.

, The recently suggested intestinal cancer stem cell marker DCLK1 is an epigenetic biomarker for colorectal cancer, Epigenetics 9(3) (2014), 346–350.

Lgr5 High /DCLK1 High phenotype is more common in early stage and intestinal subtypes of gastric carcinomas

Abstract

BACKGROUND:

OBJECTIVE:

METHOD:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Study population

Table 1 Correlation between Lgr5 and DCLK1 expressions and clinicopathological features in gastric cancer patients ( P -value; Pearson χ 2). The values shown in bold, are statistically significant

2.3 Immunohistochemistry

2.4 Scoring system

2.5 Statistical analysis

3. Results

3.1 Clinicopathological information of patients

3.3 DCLK1 expression and its association with clinicopathological parameters in gastric carcinoma

4. Discussion

5. Conclusions

Conflict of interest

Footnotes

Acknowledgments

References

Table 1
Correlation between Lgr5 and DCLK1 expressions and clinicopathological features in gastric cancer patients ( $P$ -value; Pearson $\chi$ 2). The values shown in bold, are statistically significant