Objective:To investigate the risk factors of pathological complete response(pCR)after neoadjuvant therapy for locally advanced gastric cancer(LAGC)and assess the value of gastric tumor markers for predicting pCR in LAGC patients.Methods:We retrospectively ana-lyzed the clinical and pathological characteristics of 213 patients who underwent radical gastrectomy and gastric tumor marker analysis after neoadjuvant therapy at The Chinse PLA General Hospital First Medical Center,between January 2020 and April 2024(20 and 193 cases in the pCR and non-pCR groups,respectively).The interrelationships among pCR,tumor markers,and clinicopathological features were compared,and independent risk factors for pCR were analyzed.A nomogram was constructed to predict the pCR.Results:Among 213 patients,20(9.4% )achieved pCR.Univariate analysis showed that age(P=0.067),tumor bed diameter(P<0.001),gastrin-17 levels(P=0.005),CA72-4 levels(P=0.073),pepsinogen ratio(P=0.024),and neoadjuvant immunotherapy(P=0.022)were strongly associated with pCR in LAGC pa-tients.Multivariate analysis showed that neoadjuvant immunotherapy,CA72-4 levels<2.5 U/mL,gastrin-17 levels<1.48 pmol/L,and tumor bed diameter<2.85 cm were independent predictive factors for pCR in LAGC patients(P<0.05).These indicators were incorporated into a nomogram prediction model;an receiver operating characteristic curve(ROC)was plotted with an AUC(95% CI)of 0.863(0.785-0.942).The calibration and decision curves suggested that the nomogram was well calibrated and had a good net benefit.Conclusions:Gastric tumor markers can effectively predict pCR after neoadjuvant therapy in LAGC patients.Our nomogram showed a good predictive ability for pCR.Thus,our findings can serve as a useful reference for clinical decision making for LAGC patients.

Objective To investigate the extraction methods for active components from oral ulcer film and optimize the determination methods of active components dexamethasone sodium phosphate and metronazole. Methods Different extraction solvents(methanol, water and 70% methanol aqueous) were applied to extract the active components dexamethasone sodium phosphate and metronazole from oral ulcer film, which contents were quantified by a HPLC method. Results the extraction solvent water had the best efficacy and more simpler compared to the other two solvents. Clotriazole showed a good linear relationship within 5.014 5-200.5800 μg/ml (r=0.999 8), and the average extraction recovery was (104.23±0.63)%, and for dexamethasone sodium phosphate, a good linear relationship was obtained in the range of 0.482-16.328 μg/ml (r=0.9999), and the average extraction recovery was (103.97±1.02)%. Conclusion The water extraction method established in this study was simple and efficient, which showed features of simplicity, accuracy and repeatable.

Objective:To compare short-term efficacy of robotic versus 3D laparoscopic-assisted D2 radical distal gastrectomy in gastric cancerpatients and those with different body mass index (BMI).Method:A retrospective cohort study was performed. Inclusion criteria:(1) gastric cancer proved by preoperative pathological results and tumor location was suitable for D2 radical distal gastrectomy; (2) no distal metastases such as in liver, kidney or abdominal cavity, and no direct invasion to the pancreas or colon on preoperative imaging; (3) postoperative pTNM stage ranged from I to III; (4) no conversion to open surgery or change of surgical procedure during operation; (5) complete clinicopathological data. Patients with severe chronic diseases, other malignant tumors, tumor invasion of other organs or distant metastases, benign gastric tumors, gastrointestinal stromal tumors and recurrent gastric cancer were excluded. According to the above criteria, 531 patients who underwent robotic or 3D laparoscopic-assisted distal gastrectomy at the General Surgery Department of Chinese PLA General Hospital from January 2016 to December 2019 were enrolled. Among them, 344 patients underwent 3D laparoscopic-assisted distal gastrectomy (3D-LADG group), including 250 males, 94 females, 66 cases (19.2%) with a BMI ≥ 25 kg/m 2, and 278 cases (80.8%) with a BMI < 25 kg/m 2, and 187 patients underwent robotic-assisted distal gastrectomy (RADG group), including 122 males, 65 females, 69 cases (36.9%) with a BMI≥25 kg/m 2 and 118 cases (63.1%) with a BMI < 25kg/m 2. There were no significant differences in baseline characteristics between the two groups (all P > 0.05). Operative indicators, postoperative recovery, pathological characteristics and complication rate were compared between the two groups. Subgroup analysis stratified BMI was also performed. Results:Compared with RADG group, 3D-LADG group presented more harvested lymph nodes (29.1±12.4 vs. 25.2±9.0, t=4.238, P<0.001), shorter postoperative hospital stay [8.0 (7.0 to 10.0) days vs. 10.0 (9.0 to 11.0) days, Z=-6.205, P<0.001], less operative cost [(3.6×10 4±1.1×10 4) yuan vs. (6.2×10 4±3.5×10 4) yuan, t=-9.727, P<0.001], less cost of hospitalization [8.6×10 4(7.5×10 4 to 10.0×10 4) yuan vs. 12.8×10 4(11.7×10 4 to 14.1×10 4) yuan, Z=-15.997, P<0.001] and longer first flatus time [(3.9±1.0) days vs. (3.4±1.2) days, t=4.271, P<0.001], whose differences were all statistically significant (all P<0.05). While there were no statistically significant differences in operation time, intraoperative blood loss, overall complication rate [10.8%(37/344) vs. 12.8%(24/187), χ 2=0.515, P=0.473] and severe complications rate [2.0%(7/344) vs. 3.2%(6/187), χ 2=0.294, P=0.588] between 3D-LADG group and RADG group (all P>0.05). In BMI<25 kg/m 2 group, propensity score matching (PSM) was used to reduce bias of baseline characteristics. After PSM, 3D-LADG group presented higher proportion of intraoperative blood loss <50 ml [26.7% (31/116) vs. 8.6% (10/116), χ 2=13.065, P<0.001], more harvested lymph nodes [30.3±12.2 vs. 25.3±9.5, t=-3.192, P=0.002] and shorter postoperative hospital stay [9.0 (7.0 to 10.0) days vs. 10.0 (9.0 to 11.0) days, Z=-4.275, P<0.001] compared with RADG group, while other perioperative indicators showed no statistically significant differences between the two groups (all P>0.05). In BMI≥25 kg/m 2 group, 3D-LADG group presented higher proportion of intraoperative blood loss >200 ml [18.2% (12/66) vs. 1.4% (1/69), χ 2=10.853, P=0.001] and shorter postoperative hospital stay [8.0 (6.0 to 10.0) days vs. 9.0 (8.0 to 10.5) days, Z=-3.039, P=0.002] compared with RADG group, while other perioperative indicators also showed no statistically significant differences between the two groups (all P>0.05). Conclusion:It is safe and feasible to perform 3D-LADG and RADG for patients with gastric cancer. The short-term efficacy of both is similar.

Objective:The storage of medical data has been digitized in China, but a unified and structured model has not yet been established. The standardized collection, analysis and sorting of tumor clinical data is the foundation of improving the standard of tumor diagnosis and treatment. Therefore, establishing a database platform of gastric cancer (GC) is an urgent need to integrate data resources and improve the level of diagnosis and treatment. The population economics indexes of GC patients in the last 20 years are analyzed in a single-center GC database. The medical records were structured by natural language processing technology. Authors aim to investigate the clinical pathological characteristics, staging and survival of the GC patients with gastrectomy.Method:A retrospective cohort study was carried out. Clinicopatological data of patients receiving surgical treatment from 2000 to 2019 were retrospectively collected. According to the gastric cancer TNM staging guidelines from the Union for International Cancer Control and the American Joint Committee on Cancer (UICC/AJCC) 8th edition, the structured gastric cancer clinicopathological data were re-evaluated and interpreted. The Kaplan-Meier method and the log-rank test were used to compare survival rate among different groups of patients with complete follow-up data of 2010-2016.Results:Clinicopathological data of 13 492 GC patients were enrolled. The ratio of men to women in the whole group was 3.25:1.00, including 10 320 men with average onset age of 59.68 years, which was basically stable in recent 20 years, and 3172 women with average onset age of 55.93 years, which presented a trend of average increasement of 0.17 year per year. The average hospitalization duration for GC patients showed a decreasing trend year by year, which was 13.87 days in 2019. Average hospitalization cost for GC patients was increasing year by year, with a peak of 83 600 CNY in 2017 and 75 400 CNY in 2019. By natural language identification and exclusion criteria screening, a total of 7218 GC patients obtained structured clinicopathological information. Analysis on clinicopathological characteristics of 3626 GC patients in the last 5 years showed that the average diameter of tumor was (4.44±2.61) cm; the average number of harvested lymph node was 24.30±13.29; the proportion of surgical methods were as following: open surgery in 1398 cases (38.55%), laparoscopic surgery in 1856 cases (51.19%) and robotic surgery in 372 cases (10.26%). The postoperative pathological stage was as following: IA in 658 cases (18.15%), IB in 318 cases (8.77%), IIA in 559 cases (15.42%), IIB in 543 (14.98%), III A in 632 (17.43%), III B in 612 cases (16.88%), III C in 276 cases (7.61%), and IV in 28 cases (0.77%). Complete follow-up data of 3431 patients from 2010 to 2016 were presented. The 1-, 3- and 5-year survival rates were 82%, 69% and 60%, respectively for the whole group. The 1-, 3- and 5-year survival rates for patients undergoing laparoscopic surgery were 83%, 70% and 64%, respectively, and for those undergoing open surgery were 81%, 67% and 56%, respectively, and the difference between the two groups was not statistically significant ( P=0.109). The 5-year survival rate of GC patients with different AJCC stages was as following: 88% in IA, 77% in IB, 70% in II A, 62% in II B, 44% in III A, 32% in III B, 22% in III C, and 17% in IV. Conclusion:This study provides basic data for the establishment of comprehensive diagnosis and treatment model of multicenter, shedding light on the improvement of comprehensive treatment of GC in China.

Objective:To compare short-term efficacy of robotic versus 3D laparoscopic-assisted D2 radical distal gastrectomy in gastric cancerpatients and those with different body mass index (BMI).Method:A retrospective cohort study was performed. Inclusion criteria:(1) gastric cancer proved by preoperative pathological results and tumor location was suitable for D2 radical distal gastrectomy; (2) no distal metastases such as in liver, kidney or abdominal cavity, and no direct invasion to the pancreas or colon on preoperative imaging; (3) postoperative pTNM stage ranged from I to III; (4) no conversion to open surgery or change of surgical procedure during operation; (5) complete clinicopathological data. Patients with severe chronic diseases, other malignant tumors, tumor invasion of other organs or distant metastases, benign gastric tumors, gastrointestinal stromal tumors and recurrent gastric cancer were excluded. According to the above criteria, 531 patients who underwent robotic or 3D laparoscopic-assisted distal gastrectomy at the General Surgery Department of Chinese PLA General Hospital from January 2016 to December 2019 were enrolled. Among them, 344 patients underwent 3D laparoscopic-assisted distal gastrectomy (3D-LADG group), including 250 males, 94 females, 66 cases (19.2%) with a BMI ≥ 25 kg/m 2, and 278 cases (80.8%) with a BMI < 25 kg/m 2, and 187 patients underwent robotic-assisted distal gastrectomy (RADG group), including 122 males, 65 females, 69 cases (36.9%) with a BMI≥25 kg/m 2 and 118 cases (63.1%) with a BMI < 25kg/m 2. There were no significant differences in baseline characteristics between the two groups (all P > 0.05). Operative indicators, postoperative recovery, pathological characteristics and complication rate were compared between the two groups. Subgroup analysis stratified BMI was also performed. Results:Compared with RADG group, 3D-LADG group presented more harvested lymph nodes (29.1±12.4 vs. 25.2±9.0, t=4.238, P<0.001), shorter postoperative hospital stay [8.0 (7.0 to 10.0) days vs. 10.0 (9.0 to 11.0) days, Z=-6.205, P<0.001], less operative cost [(3.6×10 4±1.1×10 4) yuan vs. (6.2×10 4±3.5×10 4) yuan, t=-9.727, P<0.001], less cost of hospitalization [8.6×10 4(7.5×10 4 to 10.0×10 4) yuan vs. 12.8×10 4(11.7×10 4 to 14.1×10 4) yuan, Z=-15.997, P<0.001] and longer first flatus time [(3.9±1.0) days vs. (3.4±1.2) days, t=4.271, P<0.001], whose differences were all statistically significant (all P<0.05). While there were no statistically significant differences in operation time, intraoperative blood loss, overall complication rate [10.8%(37/344) vs. 12.8%(24/187), χ 2=0.515, P=0.473] and severe complications rate [2.0%(7/344) vs. 3.2%(6/187), χ 2=0.294, P=0.588] between 3D-LADG group and RADG group (all P>0.05). In BMI<25 kg/m 2 group, propensity score matching (PSM) was used to reduce bias of baseline characteristics. After PSM, 3D-LADG group presented higher proportion of intraoperative blood loss <50 ml [26.7% (31/116) vs. 8.6% (10/116), χ 2=13.065, P<0.001], more harvested lymph nodes [30.3±12.2 vs. 25.3±9.5, t=-3.192, P=0.002] and shorter postoperative hospital stay [9.0 (7.0 to 10.0) days vs. 10.0 (9.0 to 11.0) days, Z=-4.275, P<0.001] compared with RADG group, while other perioperative indicators showed no statistically significant differences between the two groups (all P>0.05). In BMI≥25 kg/m 2 group, 3D-LADG group presented higher proportion of intraoperative blood loss >200 ml [18.2% (12/66) vs. 1.4% (1/69), χ 2=10.853, P=0.001] and shorter postoperative hospital stay [8.0 (6.0 to 10.0) days vs. 9.0 (8.0 to 10.5) days, Z=-3.039, P=0.002] compared with RADG group, while other perioperative indicators also showed no statistically significant differences between the two groups (all P>0.05). Conclusion:It is safe and feasible to perform 3D-LADG and RADG for patients with gastric cancer. The short-term efficacy of both is similar.

Objective:The storage of medical data has been digitized in China, but a unified and structured model has not yet been established. The standardized collection, analysis and sorting of tumor clinical data is the foundation of improving the standard of tumor diagnosis and treatment. Therefore, establishing a database platform of gastric cancer (GC) is an urgent need to integrate data resources and improve the level of diagnosis and treatment. The population economics indexes of GC patients in the last 20 years are analyzed in a single-center GC database. The medical records were structured by natural language processing technology. Authors aim to investigate the clinical pathological characteristics, staging and survival of the GC patients with gastrectomy.Method:A retrospective cohort study was carried out. Clinicopatological data of patients receiving surgical treatment from 2000 to 2019 were retrospectively collected. According to the gastric cancer TNM staging guidelines from the Union for International Cancer Control and the American Joint Committee on Cancer (UICC/AJCC) 8th edition, the structured gastric cancer clinicopathological data were re-evaluated and interpreted. The Kaplan-Meier method and the log-rank test were used to compare survival rate among different groups of patients with complete follow-up data of 2010-2016.Results:Clinicopathological data of 13 492 GC patients were enrolled. The ratio of men to women in the whole group was 3.25:1.00, including 10 320 men with average onset age of 59.68 years, which was basically stable in recent 20 years, and 3172 women with average onset age of 55.93 years, which presented a trend of average increasement of 0.17 year per year. The average hospitalization duration for GC patients showed a decreasing trend year by year, which was 13.87 days in 2019. Average hospitalization cost for GC patients was increasing year by year, with a peak of 83 600 CNY in 2017 and 75 400 CNY in 2019. By natural language identification and exclusion criteria screening, a total of 7218 GC patients obtained structured clinicopathological information. Analysis on clinicopathological characteristics of 3626 GC patients in the last 5 years showed that the average diameter of tumor was (4.44±2.61) cm; the average number of harvested lymph node was 24.30±13.29; the proportion of surgical methods were as following: open surgery in 1398 cases (38.55%), laparoscopic surgery in 1856 cases (51.19%) and robotic surgery in 372 cases (10.26%). The postoperative pathological stage was as following: IA in 658 cases (18.15%), IB in 318 cases (8.77%), IIA in 559 cases (15.42%), IIB in 543 (14.98%), III A in 632 (17.43%), III B in 612 cases (16.88%), III C in 276 cases (7.61%), and IV in 28 cases (0.77%). Complete follow-up data of 3431 patients from 2010 to 2016 were presented. The 1-, 3- and 5-year survival rates were 82%, 69% and 60%, respectively for the whole group. The 1-, 3- and 5-year survival rates for patients undergoing laparoscopic surgery were 83%, 70% and 64%, respectively, and for those undergoing open surgery were 81%, 67% and 56%, respectively, and the difference between the two groups was not statistically significant ( P=0.109). The 5-year survival rate of GC patients with different AJCC stages was as following: 88% in IA, 77% in IB, 70% in II A, 62% in II B, 44% in III A, 32% in III B, 22% in III C, and 17% in IV. Conclusion:This study provides basic data for the establishment of comprehensive diagnosis and treatment model of multicenter, shedding light on the improvement of comprehensive treatment of GC in China.

Objective To determine the angles of trochlear line(TL), antero-posterior line (APL) and posterior condylar line (PCL) with clinical epicondylar axis (CEA), analyze the variability difference in 3 axes relative to CEA. Methods The right knees in 36 patients with knee osteoarthritis (gradeⅣ) and who had underwent total knee arthroplasty were enrolled in this study, 11 male patients aged 60-81 (69.1 ± 6.3) years and 25 female patients aged 33- 85(67.7 ± 12.2) years. All of right knees were scanned using computed tomography. The angles between the CEA and each of the 3 axes (TL, PCL, APL) were measured using software. Results The angles of TL-CEA, APL-CEA and PCL-CEA was (6.10 ± 3.22)°, (85.80 ± 2.86)°and (2.70 ± 1.80)°. The F tests showed that the angel invariability between the TL and CEA the APL axis (α=0.58) or the PCL (α=0.28) for referencing the CEA had no significant differences. Conclusions The TL can be a relatively reliable reference axis to determine rotational alignment of the femoral component similar to PCL and APL.

Objective: To compare the short-term and long-term outcome between robotic gastrectomy and laparoscopic gastrectomy. Methods: The clinical data of 517 patients who had received robotic gastectomy and laparoscopic gastrectomy between December 2011 and December 2013 at Department of General Surgery, Chinese People′s Liberation Army General Hospital was collected. After propensity score matching, 70 patients in robotic gastectomy and 70 patients in laparoscopic gastectomy were identified. Perioperative outcome and overall survival were compared between the two groups using t test, χ² test, Kaplan-Meier curve and Log-rank test, respectively. Prognosis factors were analyzed by Cox′s proportional hazards regression. Results: There were comparable baseline characteristics between patients in robotic group (RG) and those in laparoscopic group (LG). The conversion rate for RG and LG were 5.7% and 4.3% respectively (P=1.000). Compared with LG, RG had similar lymph node retrieval (25.5±7.2 vs. 24.5±8.3, t=0.770, P=0.443) and less blood loss ((147.0±96.8) ml vs. (188.0±111.2) ml, t=-2.326, P=0.021). There were also similar complications (χ²=0.233, P=0.629) and severity of complications (W=70.500, P=0.053). Although there tended to be early mobility, early flatus and less hospital stay for patients in RG group, the difference between RG and LG was not statistically significant. The 3-year survival rate was 72.9% and 60.0% for patients in RG and patients in LG (P=0.578). Multivariable analysis revealed gender (HR=2.529, 95% CI: 1.042 to 6.140, P=0.040), neoadjuvant chemotherapy (HR=0.272, 95% CI: 0.104 to 0.710, P=0.008) and vascular invasion (HR=2.135, 95% CI: 1.027 to 4.438, P=0.042) were independent prognostic factors. Conclusion Compared with laparoscopic gastrectomy, robotic gastectomy could achieve similar short-term and long-term outcomes.

OBJECTIVETo analyze relationships between the tumor deposits (TD) and clinicopathologic features of gastric cancer and investigate the value of TD in staging and prognosis in gastric cancer patients.

METHODSRetrospective cohort study was conducted to evaluate the clinicopathologic data of 388 gastric cancer patients who underwent surgical procedures in Chinese PLA General Hospital between November 2011 and December 2012. Relationships between TD and clinicopathologic features were analyzed by χor Fisher exact tests. Survival curves were also generated by Kaplan-Meier method. The univariate and multivariate analysis were performed with Log-rank and COX proportional hazard model to examine the association between prognosis and TD.

RESULTSTD were observed in 67 (17.3%) of 388 gastric cancer patients, including 48 male patients (48/289, 16.6%) and 19 female patients (19/99, 19.2%). There were 40 patients (40/198, 20.2%) whose age was above 64 years old. TNM staging of positive TD patients was as follows: for pathology, there were 5 patients (5/64, 7.8%) in stage II(b, 6 patients (6/58, 10.3%) in stage III(a, 14 patients (14/75, 18.7%) in stage III(b, 30 patients (30/135, 22.2%) in stage III(c, 12 patients (12/39, 30.8%) in stage IIII( and no one in stage I(b or II(a; for T-staging, there were 2 patients (2/18, 11.1%) in stage T2, 2 patients (2/27, 7.4%) in stage T3, 36 patients (36/259, 13.9%) in stage T4a and 27 patients (27/84, 32.1%) in stage T4b; for N-stage, there were 5 patients (5/72, 6.9%) in stage N0, 6 patients (6/72, 8.3%) in stage N1, 19 patients (19/82, 23.2%) in stage N2, 27 patients (27/100, 27.0%) in stage N3a and 10 patients(10/62, 16.1%) in stage N3b; for M-stage, there were 12 patients (12/40, 30.0%) in distal metastases; for vascular invasion, there were 29 patients (29/129, 22.5%). Among positive TD patients, the number of TD >3 was found in 38 of 67 cases(56.7%). TD was associated with pTNM-stage (χ=16.898, P=0.010), T-stage (χ=17.382, P=0.001), N-stage (χ=18.080, P=0.001), M-stage (χ=5.060, P=0.036) and vascular invasion(χ=3.675, P=0.039). The median survival time of positive TD patients was significantly shorter as compared to negative TD patients (22 months vs. 32 months, χ=23.391, P=0.012). Among positive TD patients, the median survival time of patients with TD number >3 was significantly shorter as compared to those with TD number <3 (17 months vs. 25 months, χ=5.157, P=0.023). Multivariate survival analysis showed that TD number >3 was the independent risk factor of prognosis (RR=2.350, 95%CI:1.345 to 4.106, P=0.003).

CONCLUSIONSTD state is closely associated with the staging of gastric cancer and TD number >3 indicates a poor prognosis.

Aged ; China ; Cohort Studies ; Female ; Humans ; Lymphatic Metastasis ; Male ; Middle Aged ; Multivariate Analysis ; Neoplasm Invasiveness ; pathology ; Neoplasm Metastasis ; Neoplasm Staging ; methods ; statistics & numerical data ; Prognosis ; Proportional Hazards Models ; Retrospective Studies ; Risk Factors ; Stomach Neoplasms ; classification ; diagnosis ; mortality ; pathology ; Survival Rate

OBJECTIVETo investigate the clinicopathologic features and prognostic factors of gastric neuroendocrine neoplasms(gNENs).

METHODSClinicopathologic data of 104 patients with gastric neuroendocrine neoplasms admitted in Chinese PLA General Hospital between January 2000 and December 2014 were analyzed retrospectively. Tumor proliferation activity classification (G1, G2 and G3) and TNM staging were observed. The clinicopathologic features of the whole group were collected and the univariate and multivariate analysis were determined by Log-rank and Cox proportional hazard model to detect the prognosis-determining features.

RESULTSOf all the patients, 66 cases(63.5%) were neuroendocrine carcinoma, 25 cases(24.0%) were mixed adenoendocrine carcinoma and 12 cases (11.5%) were neuroendocrine tumor. For G grades, 92 cases (88.5%) were G3 grade, 8 cases(7.7%) were G2 grade and 4 cases (3.8%) were G1 grade. TNM staging results showed that stageI( was found in 6 cases (5.8%), stageII(A in 6 cases (5.8%), stageII(B in 9 cases (8.7%), stage III(A in 8 cases (7.7%), stage III(B in 55 cases (52.9%) and stageIIII( in 20 cases (19.2%). For T stage, 7 cases (6.7%) were T1, 12 cases (11.5%) were T2, 24 cases (23.1%) were T3, and 61 cases (58.7%) were T4. Lymph node metastasis occurred in 73 cases (70.2%) and distant metastasis occurred in 20 cases(19.2%). Eighty-six patients were followed up for 6 to 186 months. The median survival was 33.0 months(95% CI: 28.3 to 36.6), and 1-, 3-, and 5-year survival rates were 80%, 49% and 31%. Clinicopathologic features which were considered statistically significant on univariate analysis were selected to Cox proportional hazard model. Univariate analysis showed that risk factors of reducing survival rate included tumor size, pathological type, proliferation activity grades, and depth of invasion (all P<0.05), as well as chromogranin A expression, tumor staging, lymph node metastasis and distant metastasis(all P<0.01). The multivariate analysis showed that the stage of gNEN was the independent risk factor of the prognosis (RR=14.213, 95% CI: 1.316 to 153.524, P=0.029).

CONCLUSIONLate staging is the main clinical feature and a prognostic factor for gNENs.

Carcinoma ; diagnosis ; pathology ; Humans ; Lymphatic Metastasis ; Multivariate Analysis ; Neoplasm Staging ; Neuroendocrine Tumors ; diagnosis ; pathology ; Prognosis ; Proportional Hazards Models ; Retrospective Studies ; Risk Factors ; Stomach Neoplasms ; diagnosis ; pathology ; Survival Rate