The value of contrast-enhanced computed tomography in predicting gastric cancer recurrence and metastasis

Abstract

BACKGROUND:

The present study aimed to illustrate the clinical value of contrast-enhanced computed tomography (CECT) in predicting gastric cancer (GC) recurrence and metastasis.

METHODS:

From March 2012 to March 2014, 131 GC patients who underwent initial treatment at the Tumor Hospital of Harbin Medical University were enrolled in this study, and 64-slice spiral CT was used for CECT. CECT parameters were recorded, including CT values in the arterial and venous phases, as well as the enhancement patterns in the arterial phase. All GC patients were followed for one year. The receiver operating characteristic (ROC) curve was used to evaluate the clinical values of the CECT parameters in predicting GC recurrence and metastasis.

RESULTS:

Significantly higher CT values in the arterial and venous phases were found in patients with higher tumor node metastasis (TNM) staging, lymph node metastasis (LNM), advanced Borrmann type and postoperative chemotherapy. TNM stage I-II patients showed less significant enhancement and significant heterogeneous enhancement than TNM stage III-IV patients. Compared with patients without recurrence and metastasis, patients with recurrence and metastasis demonstrated higher CT values in the arterial and venous phases, as well as a higher proportion of significant enhancement and significant heterogeneous enhancements. For the CT value in the arterial phase, the area under the ROC curve was 0.858, with 70.3% sensitivity and 95.7% specificity, and in the venous phase, it was 0.801, with 73.0% sensitivity and 83.0% specificity.

CONCLUSION:

Our study suggests that CECT has a high accuracy rate for predicting postoperative GC recurrence and metastasis.

Keywords

Contrast-enhanced computed tomography gastric cancer recurrence metastasis clinical value

1. Introduction

Gastric cancer (GC), also known as stomach cancer, develops in the lining of the stomach [1]. According to statistics, 952,000 new GC cases occurred in 2012, accounting for approximately 6.8% of all types of human cancer, making it the fifth most common malignancy worldwide; furthermore, half of all cases occur in East Asia, primarily in China [2]. GC is a multifactorial disease caused by the interaction of genetic and environmental factors, including diet, Helicobacter pylori, heredity, age and precancerous lesions [3]. Although GC treatment consists of adjuvant chemotherapy and/or chemoradiation and surgery, surgical resection of the tumor and lymph nodes remains the only curative therapy for localized GC [4, 5]. Nevertheless, over half of radically resected GC patients relapse locally or with distant metastases [5]. Clinical studies have shown that the survival rates and survival times of GC patients are closely associated with gastric wall invasion (T staging), lymph node metastasis (LNM) (N staging) and distant organ metastasis of gastric carcinomas (M staging) [6, 7]. Therefore, it is crucially important to preoperatively determine the invasion depth in the gastric wall and lymph node involvement. Currently, efforts have been directed toward improving GC diagnosis through double contrast radiography, endoscopy and biopsy, but the three major qualitative diagnostic methods are limited to observation of the mucosal surface [8, 9, 10]. With technical improvements in imaging techniques, computed tomography (CT) has shown good performance in the preoperative evaluation of GC staging.

A CT scan, also known as X-ray CT, uses computer-processed combinations of X-ray images taken from various angles to generate cross-sectional images of specific areas of a scanned object, allowing the user to identify the target inside the object without surgery [11]. Based on different scanning modes, CT scans are classified as follows: CT plain scan, enhanced scan, CT angiography and three-dimensional (3D) CT gastrography [12, 13]. Contrast-enhanced CT (CECT) increases the detection rate and diagnostic accuracy of lesions due to its better ability to differentiate the densities of lesions and adjacent normal tissues [14]. The merit of CECT is that either tumor extension or distant metastases might be evaluated in a single examination, and in detecting metastases, CECT is superior technique compared to plain CT [15]. It has been reported that CECT was useful in performing preoperative advanced GC staging [16], while few studies have systematically explored the relationship between the clinical features, recurrence and metastasis of GC and CECT. Therefore, this study aimed to further investigate the relationship between CECT and the clinical features, recurrence and metastasis of GC.

2. Materials and methods

2.1 Ethics statement

This study was approved by the Institutional Review Board of Harbin Medical University Cancer Hospital, and written informed consent was obtained from each eligible participant.

2.2 Subjects

From March 2012 to March 2014, 131 patients (96 males and 35 females, ranging in age from 26 to 80 years [mean age, 57.81 $\pm$ 11.00 years]) who underwent initial treatment and were pathologically diagnosed with GC at Harbin Medical University Cancer Hospital were enrolled in this study. The following inclusion criteria were applied: (1) patients who received routine examinations and CECT examinations before surgery and who underwent surgical treatment within 2 weeks after the CT examination, (2) patients with good compliance, and (3) patients with complete pathological data. The following exclusion criteria were applied: (1) patients with remnant GC and (2) patients with primary malignant tumors. The pathology data included smoking, drinking, tumor node metastasis (TNM) staging, lymph node metastasis (LNM), tumor classification and post-surgery treatment.

2.3 CECT

The 64-slice multidetector-row CT (Toshiba Medical Systems, Crawley, UK) was used for the CECT sans, with a 0.8 CT interval. Before undergoing CECT, the patients routinely fasted for 4 $\sim$ 5 h and were given a certain amount of warm water to fully fill and expand their stomachs, avoiding gas. The patients were placed in the supine position before being scanned. CECT was conducted as follows: the non-ionic contrast agent (370 mg/mL Ultravist) was injected intravenously into an antecubital vein at a rate of 3.5 mL/s, and the patients were then scanned in a spiral mode, with a thickness of 5 mm and a pitch of 0.984:1. A patient’s diaphragmatic dome was used as the baseline, and the reconstruction was 5 mm thick. After the injection, patients received a three-phase CT scan. The scan delay times for the arterial, portal venous and equilibrium phases were 25 s, 75 s and 3 $\sim$ 5 min, respectively. Based on the images, the CT values (Hounsfield Units, HU) during the arterial and portal venous phases, as well as the enhancement patterns in the arterial phase, were recorded.

Table 1
Association between the clinical features of GC patients and their pre-surgery CECT values

Clinical feature n CT value in arterial phase (HU) CT value in venous phase (HU)

Age (years) $>$ 60 57 65.64 $\pm$ 15.07 89.44 $\pm$ 18.31

$\leqslant$ 60 74 62.87 $\pm$ 15.28 83.03 $\pm$ 18.61

Gender Male 96 63.56 $\pm$ 14.47 84.70 $\pm$ 16.81

Female 35 65.47 $\pm$ 17.16 88.88 $\pm$ 22.72

Smoking No 78 62.77 $\pm$ 14.13 86.00 $\pm$ 18.68

Yes 53 65.99 $\pm$ 16.58 85.56 $\pm$ 18.87

Alcohol consumption No 95 63.82 $\pm$ 15.71 85.51 $\pm$ 19.78

Yes 36 64.74 $\pm$ 13.90 86.62 $\pm$ 15.66

TNM staging I 20 55.83 $\pm$ 17.45 ${}^{\rm a}$ 70.65 $\pm$ 18.94 ${}^{\rm a}$

II 17 61.28 $\pm$ 17.12 85.91 $\pm$ 19.10 ${}^{\rm b}$

III 30 67.07 $\pm$ 13.43 ${}^{\rm b}$ 92.45 $\pm$ 16.34 ${}^{\rm b}$

IV 64 65.99 $\pm$ 13.97 ${}^{\rm b}$ 87.43 $\pm$ 17.32 ${}^{\rm b}$

Lymph node metastasis No 50 58.58 $\pm$ 13.86 ${}^{\rm a}$ 79.26 $\pm$ 19.01 ${}^{\rm a}$

Yes 81 67.46 $\pm$ 15.07 ${}^{\rm b}$ 89.87 $\pm$ 17.39 ${}^{\rm b}$

Borrmann type Borrmann1 19 55.13 $\pm$ 16.87 ${}^{\rm a}$ 70.72 $\pm$ 20.58 ${}^{\rm a}$

Borrmann2 17 62.99 $\pm$ 16.70 84.46 $\pm$ 19.10

Borrmann3 48 66.20 $\pm$ 13.85 ${}^{\rm b}$ 86.66 $\pm$ 15.99 ${}^{\rm b}$

Borrmann4 47 65.90 $\pm$ 14.36 ${}^{\rm b}$ 91.55 $\pm$ 17.43 ${}^{\rm b}$

Postoperative chemotherapy No 117 62.60 $\pm$ 14.31 ${}^{\rm a}$ 84.66 $\pm$ 18.89 ${}^{\rm a}$

Yes 14 76.36 $\pm$ 17.25 ${}^{\rm b}$ 95.53 $\pm$ 13.91 ${}^{\rm b}$

Clinical feature	n	CT value in arterial phase (HU)	CT value in venous phase (HU)
Age (years)	$>$ 60	57	65.64	$\pm$ 15.07	89.44	$\pm$ 18.31
	$\leqslant$ 60	74	62.87	$\pm$ 15.28	83.03	$\pm$ 18.61
Gender	Male	96	63.56	$\pm$ 14.47	84.70	$\pm$ 16.81
	Female	35	65.47	$\pm$ 17.16	88.88	$\pm$ 22.72
Smoking	No	78	62.77	$\pm$ 14.13	86.00	$\pm$ 18.68
	Yes	53	65.99	$\pm$ 16.58	85.56	$\pm$ 18.87
Alcohol consumption	No	95	63.82	$\pm$ 15.71	85.51	$\pm$ 19.78
	Yes	36	64.74	$\pm$ 13.90	86.62	$\pm$ 15.66
TNM staging	I	20	55.83	$\pm$ 17.45 ${}^{\rm a}$	70.65	$\pm$ 18.94 ${}^{\rm a}$
	II	17	61.28	$\pm$ 17.12	85.91	$\pm$ 19.10 ${}^{\rm b}$
	III	30	67.07	$\pm$ 13.43 ${}^{\rm b}$	92.45	$\pm$ 16.34 ${}^{\rm b}$
	IV	64	65.99	$\pm$ 13.97 ${}^{\rm b}$	87.43	$\pm$ 17.32 ${}^{\rm b}$
Lymph node metastasis	No	50	58.58	$\pm$ 13.86 ${}^{\rm a}$	79.26	$\pm$ 19.01 ${}^{\rm a}$
	Yes	81	67.46	$\pm$ 15.07 ${}^{\rm b}$	89.87	$\pm$ 17.39 ${}^{\rm b}$
Borrmann type	Borrmann1	19	55.13	$\pm$ 16.87 ${}^{\rm a}$	70.72	$\pm$ 20.58 ${}^{\rm a}$
	Borrmann2	17	62.99	$\pm$ 16.70	84.46	$\pm$ 19.10
	Borrmann3	48	66.20	$\pm$ 13.85 ${}^{\rm b}$	86.66	$\pm$ 15.99 ${}^{\rm b}$
	Borrmann4	47	65.90	$\pm$ 14.36 ${}^{\rm b}$	91.55	$\pm$ 17.43 ${}^{\rm b}$
Postoperative chemotherapy	No	117	62.60	$\pm$ 14.31 ${}^{\rm a}$	84.66	$\pm$ 18.89 ${}^{\rm a}$
	Yes	14	76.36	$\pm$ 17.25 ${}^{\rm b}$	95.53	$\pm$ 13.91 ${}^{\rm b}$

Note: GC, gastric cancer; HU, Hounsfield Units; CECT, contrast-enhanced computed tomography; TNM, tumor node metastasis; ${}^{\rm a}$ and ${}^{\rm b}$ indicate significant between-group differences, $P<$ 0.05.

2.4 Image analysis

The CECT images were analyzed by 2 (blinded) experienced doctors from either the radiology department or the nuclear medicine department, and they based their diagnoses on these images. When the diagnoses conflicted, the doctors conferred further to reach a final conclusion. The slices with the highest uptake of fluorodeoxyglucose (FDG) in focus were selected, and the region of interest (ROI) was used to delineate the focus contour. The ROI included the whole focus but avoided the focuses with hypermetabolism. The maximized standardized uptake value (SUVmax) was calculated using total body weight (TBW). The central slices of focus in the CT plain scan image and image of the stomach phase were found, and the ROI was drawn in the same region with same shape and size. The ROI included solid parts of focuses and parts with the most enhancements and avoided the cavity and necrosis. CT enhancement value $=$ the mean CT value at the stomach phase – the mean CT value of the plain scan. If the short axis of the lymph node metastasis was $\geqslant$ 10 mm in the CECT cross sectional image, then it was regarded as abnormal.

Table 2
Association of TNM staging with enhancement patterns in the arterial phase of GC patients

TNM n Slight Slight heterogeneous Significant Significant heterogeneous

staging enhancement enhancement enhancement enhancement

I 20 4 (20%) 3 (15%) 6 (30%) 7 (35%)

II 17 5 (29.4%) 3 (17.6%) 7 (41.2) 2 (11.8%)

III 30 6 (20%) 1 (3.3%) 10 (33.3%) 13 (43.3%)

IV 64 13 (20.3%) 4 (6.25%) 14 (21.9%) 33 (51.6%)

Total 131 28 (21.4%) 11 (8.4%) 37 (28.2%) 55 (42%)

TNM	n	Slight	Slight heterogeneous	Significant	Significant heterogeneous
I	20	4	(20%)	3	(15%)	6	(30%)	7	(35%)
II	17	5	(29.4%)	3	(17.6%)	7	(41.2)	2	(11.8%)
III	30	6	(20%)	1	(3.3%)	10	(33.3%)	13	(43.3%)
IV	64	13	(20.3%)	4	(6.25%)	14	(21.9%)	33	(51.6%)
Total	131	28	(21.4%)	11	(8.4%)	37	(28.2%)	55	(42%)

Note: TNM, tumor node metastasis; GC, gastric cancer.

2.5 Follow-up

All participating GC patients were followed-up for one year. After surgery, the patients were followed-up monthly. The postoperative recurrence and metastasis of GC were pathologically confirmed through surgery or gastroscopy and, eventually, recorded.

2.6 Statistical analysis

SPSS software (version 19.0, SPSS Inc., Chicago, IL, USA) was used for the statistical analysis. The measurement data are expressed as the mean $\pm$ standard deviation (SD) ( $\bar{x}$ $\pm$ s); data between the two groups were compared using the $t$ -test; the data among multiple groups were compared using one-way analysis of variance (ANOVA), and data among multiple groups were compared in pairs using the least significant difference (LSD) test. The enumeration data are expressed as percentages and comparisons between two groups were analyzed using the chi-squared test. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of CECT parameters in predicting GC recurrence and metastasis. $P<$ 0.05 (bilateral) indicated a statistical difference.

3. Results

3.1 Relationships between the clinical featuresof GC patients and CECT values before surgery

The correlation between the clinical features of GC patients and the values of pre-surgical CECT is shown in Table 1. The CT values in the arterial and venous phases showed no correlation with age, gender, smoking or alcohol consumption (all $P>$ 0.05). However, significantly higher CT values in the arterial and venous phases were found in patients with higher TNM staging, LNM, advanced Borrmann type and postoperative chemotherapy (all $P<$ 0.05).

3.2 Relationships between TNM staging and enhancement patterns in GC patients in the arterial phase

Based on the CT images of GC patients, the enhancement patterns in the arterial phase were categorized as slight enhancement ( $n=$ 28), slight heterogeneous enhancement ( $n=$ 11), significant enhancement ( $n=$ 37) and significant heterogeneous enhancement ( $n=$ 55). As shown in Table 2, 56.7% of the patients with TNM stage I $\sim$ II (21/37) showed significant enhancement and significant heterogeneous enhancement, and the proportion was much lower than that of the patients with TNM stage III $\sim$ IV (74.5%, 70/94) ( $P=$ 0.048).

3.3 Association between recurrence and metastasis and pre-surgical CECT values and certain clinical features in GC patients

The 131 patients were followed-up postoperatively for one year, during which 37 patients demonstrated recurrence and metastasis. As shown in Table 3, compared with patients without recurrence and metastasis, those with recurrence and metastasis showed higher CT values in the arterial and venous phases, as well as a higher proportion of significant enhancements and significant heterogeneous enhancements (all $P<$ 0.05). Recurrence and metastasis were associated with TNM staging, LNM, Borrmann type and postoperative chemotherapy (all $P<$ 0.05).

Table 3
Association between recurrence and metastasis and pre-surgery CECT values and certain clinical features in GC patients

Clinical feature With recurrence and Without recurrence and $P$

metastasis ( $n=$ 37, HU) metastasis ( $n=$ 94, HU)

Arterial phase CT value 78.92 $\pm$ 14.19 58.23 $\pm$ 11.06 $<$ 0.001

Venous phase CT value 99.23 $\pm$ 16.19 80.54 $\pm$ 16.94 $<$ 0.001

Enhancement pattern in the arterial phase 0.029

Slight enhancement 4 24

Slight heterogeneous enhancement 3 8

Significant enhancement 17 20

Significant heterogeneous enhancement 13 42

TNM staging 0.045

I 10 10

II 2 15

III 10 20

IV 15 49

Lymph node metastasis 0.041

No 9 41

Yes 28 53

Borrmann type 0.039

Borrmann1 3 16

Borrmann2 5 12

Borrmann3 9 39

Borrmann4 20 27

Postoperative chemotherapy 0.002

No 28 89

Yes 9 5

Clinical feature	With recurrence and	Without recurrence and	$P$
	metastasis ( $n=$ 37, HU)	metastasis ( $n=$ 94, HU)
Arterial phase CT value	78.92 $\pm$ 14.19	58.23 $\pm$ 11.06	$<$ 0.001
Venous phase CT value	99.23 $\pm$ 16.19	80.54 $\pm$ 16.94	$<$ 0.001
Enhancement pattern in the arterial phase			0.029
Slight enhancement	4	24
Slight heterogeneous enhancement	3	8
Significant enhancement	17	20
Significant heterogeneous enhancement	13	42
TNM staging			0.045
I	10	10
II	2	15
III	10	20
IV	15	49
Lymph node metastasis			0.041
No	9	41
Yes	28	53
Borrmann type			0.039
Borrmann1	3	16
Borrmann2	5	12
Borrmann3	9	39
Borrmann4	20	27
Postoperative chemotherapy			0.002
No	28	89
Yes	9	5

Note: HU, Hounsfield Units; CECT, contrast-enhanced computed tomography; TNM, tumor node metastasis; GC, gastric cancer.

3.4 Predictive value of the arterial phase CT value and the venous phase CT value in recurrence and metastasis in GC patients

The ROC curves were used to evaluate the sensitivities and specificities of the CT values in the arterial and venous phases in predicting GC recurrence and metastasis, and the results are shown in Fig. 1. Figure 1A shows the ROC curve of the predictive value of the CT value in the arterial phase, and the area under the curve (AUC) was 0.858, with 70.3% sensitivity and 95.7% specificity. As for the ROC curve of the predictive value of the CT value in the venous phase (Fig. 1B), the AUC was 0.801, with the sensitivity and specificity being 73.0% and 83.0%, respectively.

Figure 1.

The ROC curves of CT values in predicting GC recurrence and metastasis. Note: A, the ROC curve of the arterial phase CT value in predicting GC recurrence and metastasis; B, the ROC curve of the venous phase CT value in predicting GC recurrence and metastasis; ROC, receiver operating characteristic; CT, computed tomography; GC, gastric cancer.

4. Discussion

In this study, the preoperative data concerning CECT parameters and clinical features, as well as the recurrence and metastasis of 131 GC patients, were retrospectively analyzed to explore the association between CECT and recurrence and metastasis in GC patients. The results indicated that in GC patients, CECT showed higher accuracy in determining TNM staging and played an important role in predicting postoperative recurrence and metastasis.

GC is a heterogeneous and multifactorial disease, the incidence and mortality of which vary geographically, with the highest rates in East Asia [17]. However, among different therapeutic methods, surgery has been considered as the only curative treatment for GC [5]. To identify the most adequate therapy, it is necessary to accurately determine preoperative staging using CT, the standard imaging modality for preoperative staging [16]. CECT has become the preferred method for GC staging due to its high spatial resolution, tissue resolution and 3D reconstruction techniques [6]. In the present study, the clinical features of GC patients were studied, and it was found that the CT values in the arterial and venous phases were significantly and positively correlated to higher TNM staging, LNM, advanced Borrmann type and postoperative chemotherapy. Confirming these results, Ba-Ssalamah et al. have reported that textural gray-level patterns on CECT images, which reflect tumor heterogeneity, may indirectly capture the cell morphology [18], and morphology showed a close relationship to lymph node involvement [19]. Similarly, a previous study has also reported that CECT showed better diagnostic performance for primary tumors and regional LNM, as well as peritoneal seeding, exhibiting high diagnostic value for preoperative staging of advanced GC [16]. It has been reported that CECT provided useful information, such as gastrointestinal neuroendocrine carcinoma locations, density, enhancement pattern, and some metastases, further increasing the diagnostic accuracy [20]. The maximum HU value represents the highest degree of tumor enhancement; a published result has shown that the parameter was significantly higher in patients with poor prognoses, which may reflect the higher vascularity that characterizes more aggressive tumors [21].

However, it has also been found that patients with TNM stage I-II showed less significant enhancement and significant heterogeneous enhancement than those with TNM stage III or IV, which implied a high accuracy for CECT in determining TNM staging. TNM staging includes local tumor depth invasion (T), regional lymph node invasion (N), and distant metastases (M) [22]. Similar to most solid tumors, GC requires neovascularization to supply additional nutrients for growing metabolic demands and to provide potential routes for tumor dissemination and metastasis. Angiogenesis develops as the GC continues to develop, thus, the peak enhancement increases correspondingly [23]. Thus, it is reasonable that high TNM staging is closely related to increased neovascularization and significant enhancement. A recent study has demonstrated that enhanced CT imaging showed higher accuracy in predicting T3 and T4 stages of GC preoperatively [6]. Hallinan et al. have reported that texture analysis for GC could represent an additional, prognostic indicator over conventional methods, such as tumor volume, which is known to be correlated with TNM stage and prognosis [24]. In addition, Lee et al. have noted that contrast-enhanced helical CT could also be successfully used to preoperatively evaluate the staging of remnant GC in patients who have undergone gastric resection [25].

In this study, the diagnostic significance of the arterial phase CT value and the venous phase CT value in predicting GC recurrence and metastasis was confirmed due to their high accuracy rates; in addition, the AUCs of the corresponding ROC curves for the arterial and venous phases were 0.858 and 0.801, respectively. In terms of GC recurrence and metastasis, D2-40 positive lymphatic vessel density (LVD) and CD34 positive microvessel density (MVD) were two valuable parameters used to detect the invasion and metastasis of tumor cells in lymphatic and blood vessels [26]. Because the CECT values were shown to positively correlate with MVD, an increased CT value indicated increased angiogenesis, as well as a higher possibility of tumor invasion and metastasis [27, 28]. Regarding GC, a published study has proven the accuracy of multidetector-row CT in diagnosing lymph node metastasis [29].

5. Conclusion

In conclusion, our study demonstrates that CECT is highly accurate in determining TNM staging for GC patients and that it plays an important role in predicting postoperative recurrence and metastasis. Nevertheless, this study only investigated the diagnostic value of CECT in GC patients with early recurrence and metastasis. Further studies are needed to evaluate the clinical value of CECT in predicting late recurrence and metastasis.

Acknowledgments

The authors thank everyone who assisted with editing this paper.

Conflicts of interest

The authors have declared no conflicts of interest regarding the publication of this paper.

References

Brzozowa

Michalski

Harabin-Slowinska

and Wojnicz

, The role of tumour microenvironment in gastric cancer angiogenesis, Prz Gastroenterol 9 (2014), 325–328.

Ferlay

Soerjomataram

Dikshit

Eser

Mathers

Rebelo

Parkin

D.M.

Forman

and Bray

, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int J Cancer 136 (2015), E359–E386.

Compare

Rocco

and Nardone

, Risk factors in gastric cancer, Eur Rev Med Pharmacol Sci 14 (2010), 302–308.

Brenkman

H.J.

Haverkamp

Ruurda

J.P.

and van Hillegersberg

, Worldwide practice in gastric cancer surgery, World J Gastroenterol 22 (2016), 4041–4048.

Orditura

Galizia

Sforza

Gambardella

Fabozzi

Laterza

M.M.

Andreozzi

Ventriglia

Savastano

Mabilia

Lieto

Ciardiello

and De Vita

, Treatment of gastric cancer, World J Gastroenterol 20 (2014), 1635–1649.

Wang

Zhao

Liu

Zhao

and Luo

, Prediction of T stage in gastric carcinoma by enhanced CT and oral contrast-enhanced ultrasonography, World J Surg Oncol 13 (2015), 184.

Peng

C.W.

Wang

L.W.

Zeng

W.J.

Yang

X.J.

and Li

, Evaluation of the staging systems for gastric cancer, J Surg Oncol 108 (2013), 93–105.

Asaka

and Mabe

, Strategies for eliminating death from gastric cancer in Japan, Proc Jpn Acad Ser B Phys Biol Sci 90 (2014), 251–258.

Lee

S.H.

Lim

K.H.

Song

S.Y.

Lee

H.Y.

Park

S.C.

Kang

C.D.

Lee

S.J.

Choi

D.W.

Park

S.B.

and Ryu

Y.J.

, Occult gastric cancer with distant metastasis proven by random gastric biopsy, World J Gastroenterol 22 (2016), 4270–4274.

10.

Giganti

Orsenigo

Arcidiacono

P.G.

Nicoletti

Albarello

Ambrosi

Salerno

Esposito

Petrone

M.C.

Chiari

Staudacher

Del Maschio

and De Cobelli

, Preoperative locoregional staging of gastric cancer: is there a place for magnetic resonance imaging? Prospective comparison with EUS and multidetector computed tomography, Gastric Cancer 19 (2016), 216–225.

11.

Vamvakeros

Jacques

S.D.

Di Michiel

Senecal

Middelkoop

Cernik

R.J.

and Beale

A.M.

, Interlaced X-ray diffraction computed tomography, J Appl Crystallogr 49 (2016), 485-496.

12.

A.D.

Zhang

Xue

and Li

, Angiogenesis of hepatocellular carcinoma under multislice spiral CT plain scan and enhanced scan, J Biol Regul Homeost Agents 29 (2015), 895–903.

13.

Miyazaki

Kikuchi

Hiramatsu

Ozaki

Iino

Ohta

Kamiya

Sakaguchi

Unno

and Konno

, Three-dimensional fusion images combining CT gastrography and CT angiography for early gastric cancer: pilot experiences of preoperative simulation prior to totally laparoscopic gastrectomy, Asian J Endosc Surg 8 (2015), 54–58.

14.

Morimoto

Tateishi

Maeda

Arai

Nakajima

and Edmund Kim

, Nodal status of malignant lymphoma in pelvic and retroperitoneal lymphatic pathways: comparison of integrated PET/CT with or without contrast enhancement, Eur J Radiol 67 (2008), 508–513.

15.

Tanaka

Sato

Fujioka

Takahashi

Kimura

Iwamoto

and Uchiyama

, Use of contrast-enhanced computed tomography in clinical staging of asymptomatic breast cancer patients to detect asymptomatic distant metastases, Oncol Lett 3 (2012), 772–776.

16.

Kim

E.Y.

Lee

W.J.

Choi

Lee

S.J.

Choi

J.Y.

Kim

B.T.

and Kim

H.S.

, The value of PET/CT for preoperative staging of advanced gastric cancer: comparison with contrast-enhanced CT, Eur J Radiol 79 (2011), 183–188.

17.

Lin

Ueda

Kikuchi

Totsuka

Wei

W.Q.

Qiao

Y.L.

and Inoue

, Comparative epidemiology of gastric cancer between Japan and China, World J Gastroenterol 17 (2011), 4421–4428.

18.

Ba-Ssalamah

Muin

Schernthaner

Kulinna-Cosentini

Bastati

Stift

Gore

and Mayerhoefer

M.E.

, Texture-based classification of different gastric tumors at contrast-enhanced CT, Eur J Radiol 82 (2013), e537–e543.

19.

Sasaki

Kishimoto

Kawabata

Sato

and Tsuchida

, Risk factors for cervical lymph node metastasis in superficial head and neck squamous cell carcinoma, J Med Dent Sci 62 (2015), 19–24.

20.

Wang

Zhang

G.B.

Yan

Wei

X.E.

Zhang

Y.Z.

and Li

W.B.

, CT and enhanced CT in diagnosis of gastrointestinal neuroendocrine carcinomas, Abdom Imaging 37 (2012), 738–745.

21.

Giganti

Antunes

Salerno

Ambrosi

Marra

Nicoletti

Orsenigo

Chiari

Albarello

Staudacher

Esposito

Del Maschio

and De Cobelli

, Gastric cancer: texture analysis from multidetector computed tomography as a potential preoperative prognostic biomarker, Eur Radiol 27 (2017), 1831–1839.

22.

Seevaratnam

Cardoso

McGregor

Lourenco

Mahar

Sutradhar

Law

Paszat

and Coburn

, How useful is preoperative imaging for tumor, node, metastasis (TNM) staging of gastric cancer? A meta-analysis, Gastric Cancer 15 Suppl 1 (2012), S3–S18.

23.

Zheng

Takahashi

Murai

Cui

Nomoto

Niwa

Tsuneyama

and Takano

, Expressions of MMP-2, MMP-9 and VEGF are closely linked to growth, invasion, metastasis and angiogenesis of gastric carcinoma, Anticancer Res 26 (2006), 3579–3583.

24.

Hallinan

J.T.

Venkatesh

S.K.

Peter

Makmur

Yong

W.P.

and So

J.B.

, CT volumetry for gastric carcinoma: association with TNM stage, Eur Radiol 24 (2014), 3105–3114.

25.

Lee

I.J.

Lee

J.M.

Kim

S.H.

Chang

Han

J.K.

Choi

B.I.

Lee

H.J.

Yang

H.K.

and Lee

K.U.

, Helical CT evaluation of the preoperative staging of gastric cancer in the remnant stomach, AJR Am J Roentgenol 192 (2009), 902–908.

26.

Cao

Y.W.

Liu

Wang

P.F.

C.R.

and Ding

H.Z.

, Lymphangiogenic and angiogenic microvessel density in chinese patients with gastric carcinoma: correlation with clinicopathologic parameters and prognosis, Asian Pac J Cancer Prev 14 (2013), 4549–4552.

27.

Shan

Wang

Chen

Xiao

Jiang

Wang

and Fan

, Necrosis degree displayed in computed tomography images correlated with hypoxia and angiogenesis in breast cancer, J Comput Assist Tomogr 37 (2013), 22–28.

28.

Bialas

Dyduch

Dudala

Bereza-Buziak

Hubalewska-Dydejczyk

Budzynski

and Okon

, Study of microvessel density and the expression of vascular endothelial growth factors in adrenal gland pheochromocytomas, Int J Endocrinol 2014 (2014), 104129.

29.

Saito

Kurokawa

Takiguchi

Miyazaki

Takahashi

Yamasaki

Miyata

Nakajima

Mori

and Doki

, Accuracy of multidetector-row CT in diagnosing lymph node metastasis in patients with gastric cancer, Eur Radiol 25 (2015), 368–3674.

TNM	n	Slight		Slight heterogeneous		Significant		Significant heterogeneous
staging		enhancement		enhancement		enhancement		enhancement
I	20	4	(20%)	3	(15%)	6	(30%)	7	(35%)
II	17	5	(29.4%)	3	(17.6%)	7	(41.2)	2	(11.8%)
III	30	6	(20%)	1	(3.3%)	10	(33.3%)	13	(43.3%)
IV	64	13	(20.3%)	4	(6.25%)	14	(21.9%)	33	(51.6%)
Total	131	28	(21.4%)	11	(8.4%)	37	(28.2%)	55	(42%)