DOI: https://doi.org/10.21203/rs.3.rs-455688/v1
Background: The prognostic value of inflammatory response makers for predicting clinical outcome have been proved valid in various cancers.The aim of this study was to explore the influence of the neutrophil-to-lymphocyte ratio(NLR),platelet-to-lymphocyte ratio(PLR)and lymphocyte-to-monocyte ratio(LMR) on the prognosis of gastric neuroendocrine tumor.
Methods: One hundred and three patients who underwent curative surgery for gastric neuroendocrine tumor were enrolled form 2009-2018 in The first affiliated hospital of Anhui Medical University.NLR,PLR and LMR were calculated from peripheral blood cell counts taken before surgery.Optimal cutoff values of NLR,PLR and LMR were determined on the basis of receiver operating characteristic curve analysis,and their prognostic values were assessed using Kaplan-Meier curve,univariate and multivariate COX regression models.
Result:The cutoff value of NLR,PLR and LMR was 2.08,147.5 and 3.94,respectively.Elevated NLR,PLR and declined LMR were found to relate with diminished Overall survival (OS) in surgical gNET patients.Multivariate analysis identified the LMR(HR=0.923 (0.860-0.991), P=0.027);tumor size(HR=1.130 (1.032-1.239),P = 0.009) and high grade postoperative complication(HR=2.847 (1.129-7.480), P =0.027) as independent prognostic factors.
Conclusion: Inflammatory response markers can predict clinical outcome in patients with gNET,especially LMR might be considered as a convenient indicator for patients’ prognosis
Through the analysis of the data of 103 patients with gNET, we determined the predictive significance of the preoperative inflammatory indicators NLR, PLR and LMR on the patient's short-term and long-term results, which will help the clinical diagnosis and treatment of gNET patients in the future.
Gastric neuroendocrine tumors (gNETs) are neoplasms derived from the enterocharomaffin -like cells (ECL cells ) of the gastric mucosa .They are rare lesions with an indolent behavior and neuroendocrine differentiatio(1). A recent independent analysis of the SEER database also found that the incidence of gastrointestinal NETs increased from 1975 to 2008(2). The reasons for this increase are unclear, although it seems likely that improved diagnosis
and classification is a factor(3).gastric NETs consist of a complex disease that includes different subtypes with distinct management and prognosis.Surgery is still the mainstay for the treatment of local or locoregional gNETs .Patients with gNETs that are symptomatic intermediate -to-high grade,or measure greater than 2cm are recommended to undergo formal oncological surgery,and the surgical approach for gNETs primarily depends on tumor location(4).
In recent years ,thanks to a higher early diagnosis rate and a wider range of treatment options ,the Overall survival rate of patients with NETs has increased to a certain extent than before.However ,a clearer understanding of prognosis factors for OS in patients with gNETs may therefore facilitate the implementation of treatment guidelines recommending the individualiazation of therapy(3).
According to our best knowledge of the literature ,inflammation is regarded as a pivotal driver for the development and progression of cancer.The neutrophil-to-lymphocyte ratio (NLR); lymphocyte-to-monocyte ratio (LMR) and platelet-to-lymphocyte ratio (PLR) can reflect the state of tumor microenvironment composed of inflammatory factors and inflammatory cells ,and can be used to assess the degree of systematic inflammation (5–7).
Studies have proved that systematic inflammatory indicators can be used as a reliable predict tools to guide the clinical prognosis in several cancers.Preoperative NLR,PLR and LMR have been shown to be related to prognosis in a variety of cancer including :colorectal cancer ,esophagus cancer ,ovarian cancer ,lung cancer, biliary tract cancer,osteosarcoma.etc(8–13).
The aim of our study was to analyse the prognostic impact of preoperative NLR,PLR and LMR in patients with gNETs.We hypothesis that elevated NLR,PLR and declined LMR are associated with unfavorable outcomes in gNETs .
A total of 103 patients who were diagnosed with gNET and underwent radical surgery at the First Affiliated Hospital of Anhui Medical University between January 2009 and August 2018 were enrolled and followed up until December 2020.The inclusion criteria are as follows :
(1) No preoperative radiotherapy, chemotherapy, immunotherapy, molecular targeted therapy, etc.;
(2) No other tumors, no history of other malignant tumors;
(3) No history of upper abdominal surgery, and suffered radical resection of gastric cancer this time
(4) No history of immune disease before operation, no other serious diseases such as acute coronary heart disease, liver cirrhosis, chronic kidney disease, chronic blood system disease, etc. No recent infection evidence;
(5) Postoperative pathological diagnosis of gastric neuroendocrine Cancer or gastric mixed neuroendocrine carcinoma;
(6) Postoperative pathological data is complete, and follow-up information is complete.
All gNET patients were confirmed in accordance with histological evidence ,or patients without histological confirmation were excluded from the present study .This study was approved by the institution ethics commission of Anhui Medical University ,and written informed consents were obtained from all participants.The pathological staging of all patients was based on the 8th edition of UICC/AJCC TNM staging .
The clinical -pathological data were collected from medical records at the Department of Hospital Medical Record .The laboratory data (white cell ,neutrophil ,lymphocyte ,monocyte and platelet counts ,AFP,CEA,CA125,CA153,CA72-4.etc) were collected from blood routine test with each patient blood sample obtained from 7 to 9am before surgical operation which was detected by Sysmex XT-1800i Automated Hematology System.
Patients ‘follow -up examinations were performed at regular intervals (every 3 months within the first to third year ,6 months within the fourth to the fifth year ).OS were defined as the time from the operation day to the date of death or the end point time of follow-up to December 30.2020.
The Receiver Operating Characteristic (ROC) curve was used to assess the sensitivity and specificity for the 5-year OS and the largest Youden’s index was estimated to determine the optimal NLR,PLR and LMR cutoff values(14) .comparison of categorical variables was conducted through chi-square test while the Student’s T test and Mann-Whitney U test were applied to the comparison of continuous variables .Survival curve were plotted by the Kaplan-Meier method and the results acquired from log-rank test were used to judge significance. The significant predictors for OS determined through univariate analysis were evaluated through multivariate analysis using Cox proportion hazards model .All data analyses was completed by SPSS 22.0 software(IBM,USA),and P༜0.05 was considered statistically significant .
As it was performed through ROC curve(Fig. 1),the areas under the curve for NLR,PLR and LMR were 0.665(P = 0.021),0.617(P = 0.039),and 0.619(P = 0.035),respectively .The optimal cut-off values were 2.08 for NLR(sensitivity ,48.94%;specificity ,78.57%);147.5 for PLR(sensitivity ,70.21%;specificity ,57.14%)and 3.94 for LMR(sensitivity ,61.7%;specificity ,64.29%).
All patients were dividend into two groups according to their NLR,PLR and LMR in turn on the basis of the three different optimal cut-off values. After the grouping is completed, we displayed the patient’s general background characteristics, tumor-related characteristics, surgery-related characteristics, and post-operative characteristics in the table 1, and the statistical values were calculated .As we can see in the table 1,between the low -and high -NLR groups ,significant discrepancy were observed in preoperative symptoms amount (P = 0.003),tumor size (P = 0.004)
intraoperative blood loss (P = 0.009),intraoperative blood transfusion (P = 0.048),postoperative complications(P = 0.008),length of postoperative hospital stay(P = 0.006);between the low- and high-PLR groups ,significant discrepancy were observed in patients age (P = 0.024), preoperative symptoms amount (P = 0.004),CA72-4 (P = 0.017),vascular invasion (P = 0.009), intraoperative blood loss (P༜0.001);and between low- and high -LMR groups,significant discrepancy were observed in No. of lymph node metastasis (P = 0.044).
The Median survival time (MST) for all 103 patients was 19months .As it is shown in the survival diagrams(Fig. 2,figure 3༌figure 4) ,the patients in the high NLR group ,high PLR group and low LMR group had a worse prognosis than the patients in the low NLR group ,low PLR group and high LMR group.
In addition to NLR,PLR and LMR.The relationship between other prognostic factors and survival time has also been explored ,All patients were divided into two groups according to various prognostic-related indicators .In terms of continuous variables ,the mean value or median is usually selected as the cut-off value while categorical variable can be grouped directly .As it is shown in Table 2,including preoperative NLR(P = 0.006) ,PLR(P = 0.006),LMR(P = 0.006),serum CA153 level (P = 0.049),tumor size (P = 0.011),tumor pathological stage (P = 0.021),high grade postoperative complications(P༜0.001) and postoperative hospitalization days(P = 0.035) has showed discrepancy in survival time between groups which is statistically significant .
We can see from table 3,Multivariate analysis revealed that high grade postoperative complications (Hazard ratio [HR], 2.847(1.129–7.480); p = 0.027), large tumor size (HR, 1.130(1.032–1.239); P = 0.009), and low preoperative LMR (HR, 0.923(0.860–0.991); p = 0.027) were strong predictive factors for poor survival,thus to be regarded as independent unfavorable factors for OS.
Gastric neuroendocrine tumor is a rare type accounting for a minority of overall gastric cancer ,with an projected prevalence of NETs in the US population in 2014 was 171,321(2) ,it’s difficult to diagnosis in early phase because of the complex clinical performance and various biological characteristics of it. The prognosis of patients with such diseases is still not optimistic, although both AJCC and ENETS have published authoritative classification and staging standards(15, 16)
Over the past decades ,more and more evidence confirmed CRI (Cancer -related inflammation ) can promote the occurrence and development of tumors ,and help them achieve all characteristic abilities including the ability to evade immune surveillance(17).Now there are studies have shown that neutrophil plays a pivotal role in many stage of tumor progression .It promotes tumor grow through secrete a variety of cytokines such as GMCSF, TNF-α,IL-1,and IL-8(6, 18).Neutrophil infiltration in human tumor is associated with poor prognosis in patients with liver cancer ,cervical cancer and colorectal cancer (19–21).In addition ,lymphocytes play a critical role in tumor-related immune reaction ,several studies have reported that lymphocyte infiltration around tumor is associated with a good response to cytotoxic therapy and better prognosis. Lymphocyte infiltration is common in NET through immunohistochemical evaluation of CD3,CD4,CD8 and CD56(22–24).Studies have shown that platelet may increase angiogenesis and release growth factors to participate in the inflammatory response(25) .Platelets aggregate and degranulate in the tumor microenvironment ,and release platelet -derived growth factors and transforming growth factors to promote tumor cell growth(26) .
NLR,PLR and LMR can be used to reflect the balance of the body’s tumor-related inflammatory response and immune response .Previous studies have shown that high NLR,PLR and low LMR usually indicate an unfavorable prognosis for tumor patients .
A study conducted by Gaitanidis et al.on pancreatic neuroendocine tumors found that preoperative NLR༞2.3 can be regarded as an independent factor in evaluating the prognosis of patients with poor progression-free survival ,regardless of whether the patient has underwent surgery(27).A retrospective analysis contain 1028 patients with gastric cancer by Shimada et al concluded that NLR is an independent risk factor for the reduced survival rate of patients with gastric cancer(28) .A meta-analysis of 2557 pancreatic cancer patients in 10 studies by Hu RJ et al .showed that lower preoperative LMR is associated with worse clinicopathological features and poor prognosis of patients with advanced PC (pancreatic cancer ) (29). A retrospective study by Li QG, Li P, Tang D, Chen J, Wang DR. showed that perioperative results, including postoperative complications, intraoperative blood loss and intraoperative blood transfusion, etc., can affect the gastric cancer patients’ survival rate after surgery. They believe that postoperative complications lead to a prolonged period of immunosuppression, so that the remaining tumor cells proliferate in the host and continue to survive (30). A study by Kraft A suggested that tumor micrometastasis may develop rapidly in the course of short-term or long-term relative immunosuppression caused by postoperative complications (31). In addition, Nash GF studies have shown that both sepsis and blood transfusion can stimulate the release of vascular endothelial growth factor, which is one of the most effective stimulators for the metastatic growth of tumor cells (32). This is consistent with the results of this study.
In summary, gastric neuroendocrine tumors are relatively rare and highly malignant neuroendocrine tumors. The preoperative systemic inflammatory indicators NLR, PLR and LMR are related to the long-term prognosis and perioperative prognosis of patients with gastric neuroendocrine cancer. We suggest that the preoperative inflammatory indicators NLR, PLR and LMR should be included in the short-term and long-term prognosis evaluation of patients with gastric neuroendocrine cancer. In addition, as far as the current research is concerned, its limitations are obvious. First of all, dynamic monitoring of NLR ,PLR and LMR throughout the perioperative period will be more effective in predicting gNETs patients ‘ survival.Furthermore,Since this study is a single-center retrospective study with a relatively small patient cohort, these results will need to be confirmed by other large-scale prospective studies of multiple centers.
Acknowledgements
Not applicable
Funding
This work was supported by the Education Department of Anhui province fund project which item number is NO.KJ2019A0249
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Authors’ contributions
ZBB designed the research and made corrections and supplements to the main content of the manuscript .FC drafted the manuscript of the research and analyzed the data .All authors read and approved the final manuscript .
Competing interests
The authors declare that they have no competing interests
Consent for publication
Not applicable
Ethics approval and consent to participate
All procedures performed in studies involving human participants were in accordance with the ethical standards of the Ethics Committee of the First Affiliated Hospital of Anhui Medical University, and with the 1964 Declaration of Helsinki and its late amendments or comparable ethical standards
Table Ⅰ ：Comparison of various factors between groups
|
Factor |
NLR |
P value |
PLR |
P value |
LMR |
P value |
|||
|
low-NLR Group |
high-NLR Group |
low-PLR Group |
high-PLR Group |
low-LMR Group |
high-LMR Group |
||||
|
（n=35） |
（n=68） |
（n=58） |
（n=45） |
（n=54） |
（n=49） |
||||
|
Patient Characteristics |
|||||||||
|
Age |
66(60-70) |
63(59-71.75) |
0.464 |
67(61.5-71.0) |
62(58.0-69.0) |
0.024* |
66(59.75-71.25) |
63(59.0-70.50) |
0.385 |
|
Sex ratio(male:female) |
27:8 |
57:11 |
0.408 |
48:10 |
36:9 |
0.720 |
47:7 |
37:12 |
0.132 |
|
BMI(kg/m2) |
21.57±2.572 |
21.65±3.157 |
0.910 |
21.65±2.845 |
21.59±3.171 |
0.926 |
21.52±2.733 |
21.74±3.272 |
0.746 |
|
ASA score(1-4) |
2(2-3) |
2(2-2) |
0.136 |
2(2-2.25) |
2(2-2) |
0.533 |
2(2-2) |
2(2-3) |
0.184 |
|
Preoperative symptoms amount |
3(2-4) |
4(3-5) |
0.003* |
3(2-5) |
4(3-5) |
0.004* |
4(3-5) |
3(2-5) |
0.160 |
|
Loss of weight (kg) |
1.0(0-3) |
2.0(0.25-5.0) |
0.157 |
1.0(0-3.25) |
2.0(1-5.0) |
0.128 |
2.0(1-5) |
1.0(0-5) |
0.191 |
|
CEA(ng/ml) |
2.91(1.80-5.65) |
2.75(1.57-6.13) |
0.981 |
2.59(1.59-5.46) |
3.20(1.95-6.60) |
0.395 |
3.13(1.60-5.86) |
2.6(1.80-5.40) |
0.548 |
|
CA199(U/ml) (normal:abnormal) |
33:2 |
61:7 |
0.681 |
52:6 |
42:3 |
0.761 |
49:5 |
45:4 |
0.844 |
|
CA72-4(U/ml incomplete) (normal:abnormal) |
19:1 |
35:11 |
0.138 |
34:3 |
20:9 |
0.017* |
26:9 |
28:3 |
0.092 |
|
AFP(ng/ml) (normal:abnormal) |
31:4 |
57:11 |
0.725 |
52:6 |
36:9 |
0.168 |
45:9 |
43:6 |
0.525 |
|
CA125(U/ml incomplete) (normal:abnormal) |
31:1 |
53:7 |
0.319 |
48:4 |
36:4 |
0.697 |
45:4 |
39:4 |
0.847 |
|
CA153(U/ml，incomplete) |
6.75(5.26-8.16) |
6.13(3.48-6.83) |
0.074 |
6.4(4.8-7.86) |
5.9(3.76-7.0) |
0.208 |
6.4(5.1-8.2) |
5.9(4.2-6.9) |
0.187 |
|
Tumor-related characteristics |
|
|
|
|
|
|
|
|
|
|
Tumor location (cardia:others) |
29:6 |
51:17 |
0.364 |
48:10 |
32:13 |
0.159 |
38:16 |
42:7 |
0.062 |
|
Tumot size(cm) |
4.0(3.0-6.0) |
5.75(4.13-7.15) |
0.004* |
5.0(3.87-7) |
6.0(4.0-7.5) |
0.115 |
5.75(4-7.2) |
5.0(3.5-6.5) |
0.081 |
|
T staging (T2+T3:T4a+T4b) |
3:32 |
4:64 |
0.920 |
2:56 |
5:40 |
0.255 |
3:51 |
4:45 |
0.599 |
|
N staging (N0=0,N1=1,N2=2,N3=3) |
2(1-2) |
1(0-3) |
0.644 |
2(0-3) |
1(0-3) |
0.639 |
2(1-3) |
1(0-2) |
0.070 |
|
M staging (M0:M1) |
32:3 |
61:7 |
0.780 |
52:6 |
41:4 |
0.804 |
47:7 |
46:3 |
0.242 |
|
Pathology Stage (UICC,7th edition) ⅠA=1ⅠB=2 ⅡA=3ⅡB=4 ⅢA=5ⅢB=6ⅢC=7 Ⅳ=8 |
6(5-7) |
6(4-7) |
0.788 |
6（4-7） |
6（5-7） |
0.352 |
6(5-7) |
6(4-7) |
0.096 |
|
NET classification (NEC: MNEC) |
22:13 |
47:21 |
0.522 |
39:19 |
30:15 |
0.951 |
37:17 |
32:17 |
0.729 |
|
NET cell classification (Large :Small) |
32:3 |
60:8 |
0.873 |
49:9 |
43:2 |
0.138 |
48:6 |
44:5 |
0.882 |
|
Nerve invasion |
7(20%) |
8(11.8%) |
0.262 |
7（12.1%） |
8（17.8%） |
0.415 |
10(18.5%) |
5(10.2%) |
0.232 |
|
Vascular invasion |
9(25.7%) |
25(36.8%) |
0.259 |
13（22.4%） |
21（46.7%） |
0.009* |
19(35.2%) |
15(30.6%) |
0.622 |
|
Ki-67 index(incomplete) |
60%(50%-65%) |
60%(50%-71.25%) |
0.772 |
60%（45%-65%） |
60%（50%-70%） |
0.706 |
60%(47.5%-67.5%) |
60%(50%-70%) |
0.287 |
|
Number of lymph nodes metastases |
3(1-6) |
2.5(0-9) |
0.748 |
3（0-7） |
2（0-9.5） |
0.814 |
4(1-9) |
1(0-5) |
0.044* |
|
Number of lymph nodes dissected |
16.51±7.233 |
16.13±6.994 |
0.796 |
16.1±6.431 |
16.5±7.832 |
0.797 |
16.72±7.288 |
15.76±6.802 |
0.489 |
|
Lymph node metastasis rate |
26.67%(5%-36.84%) |
14.64%(0-43.03%) |
0.686 |
24.26%（0-41.25%） |
14.28%（0-41.42%） |
0.859 |
31.41%(7.14%-45%) |
12.5%(0-33.3%) |
0.059 |
|
Surgical characteristics |
|
|
|
|
|
|
|
|
|
|
Operative route |
|
|
0.827 |
|
|
0.708 |
|
|
0.886 |
|
Laparotomy |
32 |
63 |
54 |
41 |
50 |
45 |
|||
|
Laparoscopic surgery |
3 |
5 |
4 |
4 |
4 |
4 |
|||
|
Excision extentsion |
|
|
0.571 |
|
|
0.874 |
|
|
0.111 |
|
Total gastrectomy |
28 |
51 |
46 |
32 |
38 |
41 |
|||
|
Distal gastrectomy |
2 |
7 |
4 |
5 |
6 |
3 |
|||
|
Proximal gastrectomy |
1 |
1 |
1 |
1 |
0 |
2 |
|||
|
Palliative surgery |
4 |
9 |
7 |
6 |
10 |
3 |
|||
|
Reconstruction way |
|
|
0.708 |
|
|
0.639 |
|
|
0.858 |
|
Roux-en-Y |
32 |
59 |
52 |
39 |
48 |
43 |
|||
|
Others |
3 |
9 |
6 |
6 |
6 |
6 |
|||
|
Operating time(minutes) |
184.29±73.895 |
187.79±62.762 |
0.811 |
190.52±75.705 |
181.56 |
0.501 |
186.39±66.591 |
186.84±66.901 |
0.973 |
|
Combined with other organ excison |
8(22.9%) |
12(17.6%) |
0.527 |
13（22.4%） |
7（15.6%） |
0.383 |
11(20.4%) |
9(18.4%) |
0.797 |
|
Intraoperative blood loss(ml) |
30(20-50) |
50(30-150) |
0.009* |
30（20-50） |
50（50-150） |
0.001* |
50(20-150) |
50(20-50) |
0.222 |
|
Intraoperative blood transfusion |
2(5.7%) |
14(20.56%) |
0.048* |
6（10.3%） |
10（22.2%） |
0.099 |
11(20.4%) |
5(10.2%) |
0.155 |
|
Postoperative characteristics |
|
|
|
|
|
|
|
|
|
|
Postoperative complications |
3(8.6%) |
22(32.4%) |
0.008* |
11（19.0%） |
14（31.1%） |
0.154 |
17(31.5%) |
8(16.3%) |
0.073 |
|
Length of postoperative hospital stay(days) |
10(8-12) |
11(10-13) |
0.006* |
11（9-13） |
10（10-12） |
0.734 |
11(10-13) |
10(9-12) |
0.411 |
|
Postoperative anal exhaust time (days) |
4.06±0.814 |
4.10±1.394 |
0.865 |
4.11±0.838 |
4.07±1.601 |
0.876 |
4.19±1.480 |
3.98±0.863 |
0.387 |
|
Postoperative time of getting out of bed(days) |
2.68±0.806 |
3.03±1.728 |
0.303 |
2.67±0.852 |
3.18±2.003 |
0.085 |
3.07±1.882 |
2.69±0.829 |
0.192 |
|
Postoperative recurrence |
23(65.7%) |
55(80.9%) |
0.089 |
42(72.4%) |
36(80%) |
0.373 |
43(79.6%) |
35(71.4%) |
0.332 |
|
Postoperative chemotherapy |
17(48.6%) |
28(41.2%) |
0.474 |
26(44.8%) |
19(42.2%) |
0.791 |
27(50%) |
18(36.7%) |
0.175 |
Footnotes：NLR,neutrophil -to-lymphocyte ratio ;PLR,platelet -to lymphocyte ratio ;LMR,lympho-
cyte -to- monocyte ratio ;BMI ,body mass index ;ASA,American society of Anesthesiologist;
CEA,carcinoembryonic antigen ;CA19-9, carbohydrate antigen 19-9;CA72-4, carbohydrate antigen
72-4;AFP, α-fetoprotein ;CA 125, carbohydrate antigen 125,CA153, carbohydrate antigen153
UICC, Union for International Cancer Control,;Cardia,the opening into the stomach and that
part of the stomach connected to the esophagus ;NEC,neuroendocrine carcinoma ;MNEC,mixed
neuroendocrine carcinoma; Ki-67，Proliferating cell related antigen
*:P＜0.05
Table2 Survival analysis of prognostic factors at different index
|
Parameter |
Cut-off value |
No.of Patient |
MST(month) |
P value |
Systemic inflammatory response markers
|
NLR High Low LMR High Low Lymphocyte count High Low PLR High low |
2.08
3.94
1.64
147.5 |
0.023* 68 13 35 30 0.006* 49 28 54 13 0.117 40 27 63 13 0.012* 47 13 56 28 |
||
|
Cancer-related prognostic factors |
||||
|
CEA(ng/ml) High Low AFP(ng/ml) High Low CA199( U/ml) High Low CA72-4（incomplete, U/ml） High Low CA125（incomplete, U/ml） High Low CA153（incomplete, U/ml） High Low Tumor size(cm) Large Small Tumor location Cardia Others NET classification NEC MNEC NET cell classification Large Small Lymphnode metastasis Yes no Vascular invasion Yes No Nerve invasion Yes No Stage(UICC 7th.Edition) Ⅰ/Ⅱ Ⅲ/Ⅳ |
4.75
5.0
8.01
5.85
9.74
6.2
5
|
31 72
28 75
50 53
33 33
46 46
35 37
48 55
80 23
69 34
92 11
74 29
34 69
15 88
25 78 |
22 19
20 18
20 19
20 32
16 28
20 28
12 28
20 17
19 20
20 12
18 27
15 22
27 19
32 17 |
0.592
0.441
0.175
0.171
0.317
0.049*
0.011*
0.453
0.500
0.228
0.096
0.441
0.810
0.021* |
|
Host-related prognostic factors |
||||
|
Age High Low Sex ratio Male Female BMI(kg/m2) High Low ASA score(1-4) High Low Loss of weight (kg) High low Preoperative symptoms amount High Low Postoperative chemotherapy Yes No |
65
21.3
3
2
5
|
54 49
84 19
39 44
20 83
37 66
34 69
45 58 |
19 20
19 30
20 20
27 18
22 19
20 19
20 19 |
0.853
0.164
0.800
0.119
0.285
0.367
0.698 |
|
Surgery-related prognostic factors |
||||
|
Operating time (minutes) High Low Intraoperative blood loss (ml) High Low Intraoperative blood Transfusion Yes No Excision extentsion Total gastrectomy Others Length of postoperative hospital stay High Low Postoperative complications (Score≧Clavien-Dindo grade III) high low |
180
50
10 |
42 61
32 71
16 87
79 24
51 52
9 94 |
22 16
18 20
28 19
22 11
12 28
10 20 |
0.914
0.705
0.255
0.059
0.035*
<0.001* |
Footnotes：NLR,neutrophil -to-lymphocyte ratio ;PLR,platelet -to lymphocyte ratio ;LMR,lympho-
cyte -to- monocyte ratio ;BMI ,body mass index ;ASA,American society of Anesthesiologist;
CEA,carcinoembryonic antigen ;CA19-9, carbohydrate antigen 19-9;CA72-4, carbohydrate antigen
72-4;AFP, α-fetoprotein ;CA 125, carbohydrate antigen 125,CA153, carbohydrate antigen153
UICC, Union for International Cancer Control,;Cardia,the opening into the stomach and that
part of the stomach connected to the esophagus ;NEC,neuroendocrine carcinoma ;MNEC,mixed
neuroendocrine carcinoma; Clavien-Dindo grade, assesses the grade of postoperative
complications； MST，median survival time (month)
*：P＜0.05.
Table Ⅲ Univariate & Multivariate regression analyses of prognostic factors
|
Parameters |
Univariate analyses |
|
Multivariate analyses |
||
|
Uncorrected HR & 95% CI |
P value |
|
Corrected HR & 95% CI |
P value |
|
|
Age |
0.994 (0.972～1.016) |
0.585 |
|
/ |
/ |
|
Sex |
|
|
|
|
|
|
male |
1.470 (0.843～2.564) |
0.174 |
|
/ |
/ |
|
female |
1 |
|
|
|
|
|
Excision extentsion |
|
|
|
|
|
|
others |
1.571 (0.970～2.545) |
0.066 |
|
1.314 (0.676～2.555) |
0.421 |
|
Total gastrectomy |
1 |
|
|
|
|
|
Postoperative complications (Score ≥Clavien-Dindo grade III) |
|
|
|
|
|
|
yes |
2.651 (1.310～5.363) |
0.007* |
|
2.847 (1.129～7.480) |
0.027* |
|
no |
1 |
|
|
|
|
|
PTM stage |
|
|
|
|
|
|
ⅡB |
1.228 (0.284～5.317) |
0.784 |
|
1.480 (0.307～7.141) |
0.626 |
|
ⅢA |
2.233 (0.493～10.110) |
0.297 |
|
2.082 (0.380～11.405) |
0.398 |
|
ⅢB |
2.009 (0.474～8.521) |
0.344 |
|
2.207 (0.428～11.384) |
0.344 |
|
ⅢC |
1.919 (0.456～8.072) |
0.374 |
|
1.258 (0.191～8.306) |
0.811 |
|
Ⅳ |
5.115 (1.092～23.961) |
0.038* |
|
2.645 (0.379～18.477) |
0.327 |
|
ⅠB |
1 |
|
|
|
|
|
High Loss weight(kg) |
0.946 (0.870～1.030) |
0.200 |
|
/ |
/ |
|
Large Tumor size(cm) |
1.118 (1.045～1.198) |
0.001* |
|
1.130 (1.032～1.239) |
0.009* |
|
High CA199(U/ml) |
1.002 (1.000～1.003) |
0.015* |
|
1.001 (0.999～1.002) |
0.336 |
|
High CA125(U/ml) |
1.009 (1.002～1.016) |
0.013* |
|
1.005 (0.996～1.014) |
0.260 |
|
Intraoperative blood transfusion |
|
|
|
|
|
|
yes |
0.716 (0.397～1.291) |
0.266 |
|
/ |
/ |
|
no |
1 |
|
|
|
|
|
High Lymphnode metastasis rate |
2.497 (1.100～5.669) |
0.029* |
|
4.170 (0.669～25.987) |
0.126 |
|
High NLR |
1.185 (1.023～1.373) |
0.024* |
|
0.948 (0.734～1.225) |
0.682 |
|
High PLR |
1.002 (0.999～1.005) |
0.161 |
|
1.002 (0.997～1.007) |
0.385 |
|
High LMR |
0.898 (0.825～0.977) |
0.013* |
|
0.923 (0.860～0.991) |
0.027* |
|
Long postoperative hospital stay |
1.042 (0.992～1.095) |
0.104 |
|
1.027 (0.967～1.091) |
0.383 |
Footnotes: CI, confidence interval; NLR,neutrophil -to-lymphocyte ratio ;PLR,platelet -to lymphocyte ratio ;LMR,lympho-cyte -to- monocyte ratio ; CA19-9, carbohydrate antigen 19-9; CA 125, carbohydrate antigen 125, Clavien-Dindo grade, assesses the grade of postoperative complications；
*: p < 0.05.