The characteristics of the included studies:
Based on the search strategy mentioned above, a total of 1110 articles were searched from PubMed, Embase and Web of Science. Finally, 41 studies published between 1994 and 2019 with a total of 28,973 patients were included in our meta-analysis according to the inclusion and exclusion criteria (6-10, 20-55). The specific literature selection process is shown in Figure 1.
Table 1 summarizes the characteristics of all the studies included in this meta-analysis. All of the included studies were retrospective. Among these studies, 23 studies were from China (7, 9, 10, 30-49), 7 studies were from the USA (6, 50-55), 7 studies were from Japan (8, 24-29), and 1 study each was from Italy (22), Australia (20), the Netherlands (21), and Turkey (23); the participants in 39 studies were patients with SCC. Three of the 41 studies included patients who received chemoradiotherapy only (26, 48, 53), and the remaining 38 studies included patients who underwent surgical resection with or without chemoradiotherapy. Among the 41 studies, 35 studies reported the relationship between the tumor length of EC and OS, 10 studies reported disease-free survival (DFS), 2 reported progression-free survival (PFS) (48, 53), 1 reported cancer-specific survival (CSS) (55) and 1 reported disease-specific survival (DSS) (20). The cut-off value (cm) applied in all studies was between 1.5 and 6. Eighteen studies used a tumor length cut-off value of ≥5, and 23 studies used a tumor length cut-off value of < 5. All the tumor length values in the Surg and Surg² groups were measured from postoperative pathological specimens, and the data from the chemoradiotherapy (CRT) group and Surg³ group were measured by endoscopy or computed tomography (CT) before treatment. The Surg¹ group did not clearly describe the timing of measurement according to the different treatments. Two studies reported only relative risk (RR) data (41, 42), and we used RR to replace HR when pooling the data.
Table1: Main characteristics of all included studies.
Author
|
Publication year
|
Patient source
|
No. of patients
|
Age(years)(median and range)
|
Follow-up(months)(median and range)
|
Treatment
|
LN
metastasis (
% of total)
|
Histology
|
Tumor location
|
TNM stage
|
Ⅲ Ⅳ(% of total)
|
Cut-off value(cm)
|
Survival analusis
|
HR
reported
|
NOS
|
Upper
|
Middle
|
Lower
|
Tachibana, M.
|
1999
|
Japan
|
65
|
65
|
NR
|
Surg²
|
30.77
|
SCC
|
6
|
59
|
T1-2N±
|
NR
|
4
|
OS
|
MV
|
7
|
Shimada Hideaki
|
2004
|
Japan
|
374
|
65
|
NR
|
Surg²
|
65.78
|
SCC
|
98
|
276
|
Ⅰ-Ⅳ
|
77.0
|
5
|
OS
|
MV
|
6
|
Barbour, A. P.
|
2008
|
Australia
|
131
|
61(30-78)
|
61
|
Surg³
|
51.15
|
SCC,AC
|
4
|
28
|
85
|
0-Ⅲ
|
32.8
|
6
|
DSS
|
MV
|
8
|
Yendamuri , S.
|
2009
|
US
|
209
|
64(33-84)
|
NR
|
Surg²
|
45.45
|
SCC,AC
|
3
|
27
|
179
|
Ⅰ-Ⅲ
|
34.9
|
3
|
OS
|
UV/MV
|
7
|
Shitara Kohei
|
2010
|
Japan
|
363
|
63
|
67.2(25.2-94.8)
|
Surg¹
|
NR
|
SCC
|
NR
|
Ⅰ-Ⅳ
|
56.2
|
5
|
OS
|
UV/MV
|
7
|
Heijl, M.V
|
2010
|
The Netherlands
|
199
|
64(35-78)
|
60+
|
Surg²
|
72.36
|
SCC,AC
|
NR
|
NR
|
NR
|
4
|
OS/DFS
|
UV/MV
|
8
|
Gaur, P.
|
2011
|
US
|
164
|
65(26-84)
|
NR
|
Surg²
|
14.63
|
AC
|
0
|
8
|
156
|
NR
|
NR
|
2
|
OS
|
UV/MV
|
8
|
Lu C. L.
|
2011
|
China
|
127
|
59(39-77)
|
NR
|
Surg²
|
32.28
|
SCC
|
NR
|
Ⅰ-Ⅲ
|
30.7
|
2
|
OS/DFS
|
MV(RR)
|
7
|
Wang Bing-Yen
|
2011
|
China
|
582
|
65.4(30-88)
|
31.9
|
Surg²
|
54.64
|
SCC
|
77
|
340
|
165
|
T1-4N0-3
|
NR
|
3
|
OS
|
MV
|
7
|
Yamamoto Sachiko
|
2011
|
Japan
|
170
|
64
|
NR
|
Surg¹
|
NR
|
SCC
|
18
|
107
|
45
|
Ⅰ
|
NR
|
5
|
OS
|
MV
|
6
|
Song Zhengbo
|
2012
|
China
|
201
|
59(31-78)
|
52(30-136)
|
Surg
|
NR
|
SCC
|
12
|
102
|
87
|
T₁-₂
|
NR
|
3
|
OS
|
MV
|
7
|
Chen, J.
|
2012
|
China
|
945
|
NR
|
NR
|
Surg²
|
100
|
SCC
|
146
|
712
|
87
|
T1-4N+
|
NR
|
5
|
OS
|
UV/MV
|
7
|
Feng, J. F.
|
2013
|
China
|
132
|
73.6 ± 2.6
|
NR
|
Surg²
|
56.11
|
SCC
|
6
|
55
|
71
|
T1-4N0-3
|
NR
|
4
|
OS
|
UV/MV
|
6
|
Matsumoto , S.
|
2013
|
Japan
|
32
|
63
|
NR
|
Surg²
|
42.86
|
SCC
|
3
|
20
|
6
|
Ⅰ-Ⅲ
|
71.9
|
6
|
OS
|
MV
|
7
|
Zeybek Arife
|
2013
|
Turkey
|
116
|
60.0 (33-75)
|
39.7
|
Surg²
|
65.52
|
SCC,AC
|
10
|
36
|
70
|
Ⅱ-Ⅲ
|
64.7
|
3
|
DFS
|
MV
|
6
|
Chen Lujun
|
2014
|
China
|
103
|
58
|
NR
|
Surg²
|
47.57
|
SCC
|
NR
|
Ⅰ-Ⅳ
|
33.0
|
3.5
|
OS
|
UV/MV
|
7
|
Shridhar Ravi
|
2014
|
US
|
154
|
65
|
NR
|
Surg¹
|
84.42
|
AC
|
0
|
5
|
80
|
Ⅰ-Ⅳ
|
68.8
|
5
|
OS
|
MV
|
6
|
Freilich, J.
|
2015
|
US
|
232
|
64.4±11.3
|
25.9(2.5-124.2)
|
Surg¹
|
83.19
|
SCC,AC
|
8
|
17
|
118
|
Ⅰ-Ⅳ
|
72.8
|
5
|
OS
|
MV
|
7
|
Hulshoff, J. B.
|
2015
|
US
|
105
|
64(57-69)
|
29(15.–56.0)
|
Surg¹
|
73.33
|
SCC,AC
|
0
|
12
|
49
|
NR
|
NR
|
5
|
DFS
|
UV/MV
|
6
|
Ma Ming-Quan
|
2015
|
China
|
362
|
54.5
|
84(6-144)
|
Surg²
|
12.18
|
SCC
|
28
|
243
|
100
|
Ⅰ-Ⅲ
|
25.1
|
4
|
OS
|
MV(RR)
|
8
|
Miao Longsheng
|
2015
|
China
|
1342
|
59.5±7.9
|
30
|
Surg²
|
45.98
|
SCC,AC
|
NR
|
Ⅰ-Ⅳ
|
41.1
|
4
|
OS
|
UV/MV
|
7
|
Hirahara Noriyuki
|
2016
|
Japan
|
147
|
65.7
|
NR
|
Surg¹
|
59.86
|
SCC
|
8
|
65
|
52
|
Ⅰ-Ⅲ
|
37.4
|
3
|
CSS
|
UV/MV
|
7
|
Hwang Jwu-Yun
|
2016
|
China
|
294
|
55
|
20.4
|
Surg²
|
58.84
|
SCC
|
51
|
114
|
129
|
Ⅰ-Ⅳ
|
49.3
|
3.2
|
OS
|
UV/MV
|
7
|
Jia Wei
|
2016
|
China
|
83
|
NR
|
NR
|
Surg²
|
56.63
|
SCC
|
8
|
44
|
31
|
Ⅰ-Ⅲ
|
43.4
|
5
|
OS/DFS
|
MV
|
6
|
Ma Qilong
|
2016
|
China
|
725
|
58(32-80)
|
NR
|
Surg²
|
46.48
|
SCC
|
NR
|
Ⅰ-Ⅲ
|
38.2
|
5
|
OS
|
UV/MV
|
7
|
Sakanaka
|
2016
|
Japan
|
144
|
67(41-85)
|
48(13-88)
|
CRT
|
74.31
|
SCC
|
33
|
79
|
32
|
Ⅰ-Ⅳ
|
63.9
|
4
|
OS
|
MV
|
7
|
Valmasoni Michele
|
2016
|
Italy
|
357
|
62 ± 9.3
|
NR
|
Surg²
|
50.98
|
SCC
|
90
|
147
|
120
|
Ⅰ-Ⅲ
|
47.6
|
3
|
OS
|
MV
|
7
|
Valmasoni Michele
|
2016
|
Italy
|
305
|
63 ± 11.2
|
NR
|
Surg²
|
66.23
|
AC
|
0
|
6
|
299
|
Ⅰ-Ⅲ
|
58.4
|
3
|
OS
|
MV
|
7
|
Wu Jie
|
2016
|
China
|
1435
|
58.3
|
24(1-128)
|
Surg²
|
53.24
|
SCC
|
33
|
691
|
711
|
T1-4N0-3M0-1
|
NR
|
4
|
OS
|
UV/MV
|
7
|
Duan, J.
|
2016
|
China
|
328
|
61
|
44.9(3.4–107.6)
|
Surg²
|
42.4
|
SCC
|
24
|
202
|
102
|
Ⅰ-Ⅲ
|
57.6
|
4.2
|
OS/DFS
|
UV/MV
|
8
|
Gao, S. H.
|
2016
|
China
|
126
|
NR
|
NR
|
Surg¹
|
85
|
SCC
|
NR
|
Ⅰ-Ⅲ
|
54.8
|
4
|
OS/DFS
|
MV
|
6
|
Li, S. P.
|
2016
|
China
|
100
|
59.2±10.3
|
NR
|
Surg¹
|
55
|
SCC
|
12
|
41
|
37
|
Ⅰ-Ⅲ
|
54.0
|
4
|
OS/DFS
|
MV
|
6
|
Tian, R.
|
2016
|
China
|
442
|
60(20-88)
|
NR
|
Surg²
|
47.5
|
SCC
|
39
|
277
|
126
|
Ⅰ-Ⅲ
|
43.7
|
5
|
OS/DFS
|
UV/MV
|
7
|
Xi Mian
|
2017
|
US
|
496
|
67(20-92)
|
24.2(2.8-155.9)
|
CRT
|
70.97
|
SCC,AC
|
138
|
358
|
Ⅰ-Ⅲ
|
67.9
|
5
|
PFS
|
MV
|
7
|
Zeng, Y.
|
2017
|
US
|
1242
|
NR
|
NR
|
Surg
|
NR
|
SCC,AC
|
39
|
136
|
913
|
NR
|
NR
|
1.5
|
OS/CSS
|
MV
|
7
|
Li, J.
|
2017
|
China
|
294
|
58(38-70)
|
26
|
Surg²
|
67.3
|
SCC
|
6
|
139
|
149
|
T1-4N±
|
NR
|
5
|
OS
|
MV
|
7
|
Yang, Y. S.
|
2017
|
China
|
508
|
59
|
37.5(1-105)
|
Surg²
|
40.9
|
SCC
|
78
|
260
|
170
|
T1-4N±
|
NR
|
4
|
OS
|
UV/MV
|
7
|
Zhang, Xiangwei
|
2017
|
China
|
498
|
59(38-81)
|
47.2(6.–64.5)
|
Surg²
|
41
|
SCC
|
NR
|
T1-4N±
|
NR
|
3
|
OS
|
UV/MV
|
7
|
Bai G.
|
2018
|
China
|
80
|
54.7±12.6
|
28(9-62)
|
Surg²
|
52.5
|
SCC
|
24
|
36
|
20
|
T1-4N±
|
NR
|
5
|
OS/DFS
|
MV
|
6
|
Cheng Ya-Fu
|
2018
|
China
|
14394
|
NR
|
NR
|
Surg¹
|
73.5
|
SCC
|
2897
|
5028
|
3213
|
0-Ⅳ
|
71.5
|
5
|
OS
|
MV
|
6
|
Xu Hong-Yao
|
2018
|
China
|
751
|
65
|
56.6(25.2-112.5)
|
CRT
|
78.96
|
SCC+Others
|
157
|
420
|
128
|
T1-4N0-3
|
NR
|
5
|
PFS
|
UV/MV
|
7
|
Gu, L.
|
2019
|
China
|
116
|
NR
|
NR
|
Surg¹
|
85.34
|
SCC
|
NR
|
Ⅰ-Ⅳ
|
55.2
|
5
|
OS
|
UV/MV
|
6
|
NO=number, HR=hazard ratio, “M” means the HR come from multivariate analysis, “U” means the HR comes from univariate analysis, NOS=Newcastle –Ottawa Quality Assessment Scale, R=Retrospective , SCC=squamous cell carcinoma , AC=adenocarcinoma , OS=overall survival, PFS=progression -free survival, DFS=disease-free survival, DSS=disease-specific survival, CSS=cancer-specific survival, NR=not reported , CRT=chemoradiation therapy, Surg¹:±Neo CRT/±Surg/±Adj CRT; Surg²: Surg±Adj CRT; Surg³: Neo CRT+Surg
|
Tumor length and OS in EC:
A total of 35 studies evaluated the relationship between tumor length and OS in EC, and because of the significant heterogeneity among the included studies (I²= 66.8%; ph < .001), the pooled HR and 95% CI were calculated by a random-effects model. The pooled HR of 1.301 (95% CI: 1.210-1.399) suggests that long tumors are associated with poor OS (Fig. 2).
Tumor length and DFS in EC:
Ten studies provided the HR and 95% CI of tumor length in association with DFS. The pooled data showed that long tumors were associated with poor DFS (HR= 1.378; 95% CI: 1.179-1.609, p <.001). Due to obvious heterogeneity (I²= 76.7%, ph < .001), the random-effects model was used to calculate the pooled HR and its 95% CI (Fig. 3).
tumor length and PFS in EC:
No significant heterogeneity was found among the studies that evaluated PFS, so a fixed-effects model was used to calculate the pooled HR and 95% CI (HR = 1.161; 95% CI: 1.008-1.337, p <.005). The data suggest that long tumors were associated with shorter PFS.
Tumor length and CSS/DSS in EC:
Two studies reported preoperative tumor length and CSS/DSS data for EC. There was no significant heterogeneity (I²= 0.0%, ph = .551). The pooled HR and 95% CI (HR=1.856; 95% CI: 1.173-2.937, p <.001) indicated that long tumors were also related to poor CSS/DSS.
Subgroup analysis:
To analyse the impact of heterogeneity on the results of this study, we performed subgroup analyses on the extracted data. Subgroup analyses by patient source (China and others), histology (SCC, AC and mixed), treatment (Surg² and others), median age (<60 and ≥ 60), cut-off value (<5 and ≥5), sample size (<200 and ≥ 200), percentage of patients with LN metastasis (<50% and ≥50%), percentage of patients with TNM stage III/IV classification (<50% and ≥50%) and HR analysis method (MV and UV) were performed to explore the potential sources of heterogeneity for the pooled OS results. Because the cut-off values of tumor length were different among the included studies, we performed subgroup analysis according to different cut-off values. Because 40% of articles chose 5 cm as the cut-off value for tumor length, we used this cut-off value as the boundary. For cut-off value ≥ 5, the pooled HR was 1.259 (95% CI: 1.096-1.446, I²=59.4%, ph = .002). For cut-off value < 5, the pooled HR was 1.322 (95% CI: 1.210-1.443, I²=67.9%, ph < .001). This suggests that we may be able to define 5 cm as a standard cut-off value and recommend it to other researchers to reduce heterogeneity between different studies. The results of subgroup analyses showed that the different classification methods had no obvious influence on HR (Table 2). In almost all subgroups, long tumors were significantly related to poor OS, which showed that our pooled HR result for OS was stable and reliable. Considering the limited research on tumor length and DFS, PFS, DSS, and CSS, no other subgroup analysis was performed.
Table 2: Subgroup analyses reflecting the association between tumor length and OS in EC patients.
|
Random-effects model
|
|
|
Fixed-effects model
|
|
|
|
Subgroup
|
No.of studies
|
HR (95% CI)
|
P
|
|
HR (95% CI)
|
P
|
I²(%)
|
Ph
|
Overall
|
35
|
1.301(1.210-1.399)
|
<.001
|
|
1.151(1.117-1.185)
|
<.001
|
66.8%
|
<.001
|
Patient source
|
|
|
|
|
|
|
|
|
China
|
23
|
1.290(1.191-1.397)
|
<.001
|
|
1.155(1.117-1.195)
|
<.001
|
64.6%
|
<.001
|
Japan
|
6
|
1.540(1.108-2.142)
|
.010
|
|
1.446(1.174-1.781)
|
<.001
|
47.0%
|
.093
|
US
|
5
|
1.457(0.851-2.495)
|
.170
|
|
1.288(1.046-1.586)
|
.017
|
84.1%
|
<.001
|
The Netherlands
|
1
|
1.080(1.008-1.157)
|
.029
|
|
1.080(1.008-1.157)
|
.029
|
—
|
—
|
Italy
|
1
|
1.258(0.890-1.778)
|
.193
|
|
1.283(1.001-1.644)
|
.049
|
—
|
—
|
Histology
|
|
|
|
|
|
|
|
|
SCC
|
28
|
1.332(1.227-1.445)
|
<.001
|
|
1.163(1.125-1.203)
|
<.001
|
64.0%
|
<.001
|
AC
|
3
|
1.616(0.950-2.748)
|
.077
|
|
1.469(1.103-1.957)
|
.009
|
67.9%
|
.044
|
mixed
|
5
|
1.128(0.895-1.421)
|
.307
|
|
1.095(1.029-1.166)
|
.004
|
78.5%
|
.001
|
Treatment
|
|
|
|
|
|
|
|
|
Surg²
|
24
|
1.311(1.207-1.423)
|
<.001
|
|
1.127(1.091-1.165)
|
<.001
|
67.3%
|
<.001
|
Others
|
11
|
1.270(1.066-1.514)
|
.007
|
|
1.250(1.170-1.336)
|
<.001
|
59.5%
|
.006
|
Sample size
|
|
|
|
|
|
|
|
|
≤200
|
16
|
1.458(1.245-1.707)
|
<.001
|
|
1.163(1.098-1.233)
|
<.001
|
56.3%
|
.003
|
>200
|
19
|
1.264(1.157-1.380)
|
<.001
|
|
1.146(1.108-1.186)
|
<.001
|
73.3%
|
<.001
|
Median age
|
|
|
|
|
|
|
|
|
<60
|
9
|
1.422(1.228-1.647)
|
<.001
|
|
1.124(1.071-1.179)
|
<.001
|
74.5%
|
<.001
|
≥60
|
14
|
1.281(1.132-1.450)
|
<.001
|
|
1.133(1.083-1.185)
|
<.001
|
67.5%
|
<.001
|
Cut-off value
|
|
|
|
|
|
|
|
|
<5cm
|
21
|
1.322(1.210-1.443)
|
<.001
|
|
1.123(1.086-1.162)
|
<.001
|
67.9%
|
<.001
|
≥5cm
|
14
|
1.259(1.096-1.446)
|
.001
|
|
1.241(1.169-1.317)
|
<.001
|
59.4%
|
.002
|
LN metastasis(% of total)
|
|
|
|
|
|
|
|
|
<50%
|
13
|
1.351(1.189-1.537)
|
<.001
|
|
1.106(1.061-1.153)
|
<.001
|
73.9%
|
<.001
|
≥50%
|
18
|
1.281(1.162-1.413)
|
<.001
|
|
1.194(1.144-1.246)
|
<.001
|
62.3%
|
<.001
|
TNM stage ⅢⅣ(% of total)
|
|
|
|
|
|
|
|
|
<50%
|
12
|
1.441(1.240-1.676)
|
<.001
|
|
1.377(1.245-1.524)
|
<.001
|
44.5%
|
.062
|
≥50%
|
14
|
1.243(1.088-1.421)
|
.001
|
|
1.190(1.134-1.249)
|
<.001
|
69.5%
|
<.001
|
HR estimate
|
|
|
|
|
|
|
|
|
UV
|
16
|
1.676(1.469-1.914)
|
<.001
|
|
1.218(1.180-1.256)
|
<.001
|
91.3%
|
<.001
|
MV
|
35
|
1.301(1.210-1.399)
|
<.001
|
|
1.151(1.117-1.185)
|
<.001
|
66.8%
|
<.001
|
HR = hazard ratio, CI = confidence interval, Ph = P value of Q test for heterogeneity test, SCC = squamous cell carcinoma, AC = adenocarcinoma; UVA = univariate analysis, MVA = multivariate analysis, LN= lymph node, Surg²: Surg±Adj CRT
|
Sensitivity analyses:
After omitting each individual study for the sensitivity analysis, the results showed that when excluding any study, the pooled HRs did not change substantially, indicating that our results were stable (Fig. 4, 5).
Publication bias:
Publication bias was evaluated by Begg’s funnel plot and Egger’s linear regression test. Considering the limited number of studies on the relationship between tumor length and DFS, PFS, DSS or CSS, we analysed the publication bias of tumor length and OS (Pr> |z| =0.127 for Begg’s test and p < .01 for Egger’s test) (Fig. 6). Therefore, we conducted a further trim-and-fill analysis (Fig. 7). The adjusted result (HR=1.195, 95% CI: 1.111-1.284) was similar to our previous pooled result. This result showed that despite the presence of publication bias, it did not substantially affect the results of our study.