Objective To explore the expression of THBS 2 and its clinical significance in gastric cancer . Methods The mRNA and protein expression levels of THBS2 were assessed in 14 paired of gastric cancer specimen and corre-sponding normal mucosa using quantitative real-time PCR and Western blot analysis. Immunohistochemistry of TH-BS2 on a population-based tissue microarray consisting of 129 gastric cancer cases was used for evaluating the TH-BS2 prognostic significance. Kaplan-Meier method and Cox′s proportional hazards model were used in survival anal-ysis. Results Both mRNA and protein expression of THBS2 in 12 gastric cancer tissues were remarkably lower than the corresponding normal tissues among total 14 pair tissues. Consistent with the results of our Western blot, THBS2 expression was significantly inhibited in gastric cancer tissues as that in the normal controls in TMA ( P=0. 031) . Overexpression of THBS2 had a significant correlation with favorable prognosis of gastric cancer patients ( P=0. 002 ) and decreasing THBS2 expression was associated with the poor histological grade of gastric cancer his-tological grade ( P=0. 005 ) . Conclusion This study suggests THBS2 is down-regulation in gastric cancer tissue versus normal gastric tissue, and it may play a critical role in gastric cancer carcinogenesis and may be a potential prognosis predictor of gastric cancer.

OBJECTIVETo explore the expression of KIF18A gene protein in gastric cancer tissues and its association with the prognosis of patients.

METHODSTwenty fresh paired gastric cancer specimens and adjacent normal mucosa(at least 5 cm from the edge of tumor) from 20 gastric cancer patients undergoing operation in Department of General Surgery at the First Affiliated Hosptial of Anhui Medical University between March 2015 and July 2015 were collected. Real-time PCR was used to examine KIF18A mRNA expression in above specimens. Meanwhile, paraffin embedded cancer tissue samples from 129 gastric cancer patients undergoing operation and 23 samples of randomly selected normal gastric tissue(adjacent non-cancer tissue) were collected to establish the microarray. Immunohistochemistry method was applied to detect the KIF18A protein expression in the microarray after confirmation by pathologists. Association of KIF18A expression with clinicopathological features in gastric cancer patients was evaluated. Cox proportional hazard model was used to identify prognostic risk factors.

RESULTSAmong 20 fresh paired gastric cancer specimens, mRNA expression of KIF18A in 16 specimens was obviously lower than that in adjacent normal tissues. The positive rate of KIF18A protein expression in gastric cancer tissues was significantly lower than that in normal gastric tissues in microarray[45.0%(58/129) vs. 69.6%(16/23), P=0.041]. KIF18A protein expression was significantly associated with invasion depth (P=0.008) and TNM staging (P=0.032). The median overall survival of all the 129 patients was 44.0(95% CI: 39.78-49.24) months. The three-year survival rates of patients with high and low KIF18A expression were 67.2% and 36.6% respectively(P=0.020). Cox regression analysis showed that KIF18A expression was an independent protective factor of the prognosis of gastric cancer patients (HR=0.570, 95% CI:0.335 to 0.970).

CONCLUSIONSKIF18A expression is down-regulated in gastric cancer tissue, which may play a critical role in gastric cancer carcinogenesis. Lower expression of KIF18A is associated with poor prognosis of gastric cancer patients. KIF18A may be a potential prognostic marker of gastric cancer.

Biomarkers, Tumor ; metabolism ; Humans ; Immunohistochemistry ; Kinesin ; metabolism ; Neoplasm Staging ; Prognosis ; Proportional Hazards Models ; Real-Time Polymerase Chain Reaction ; Regression Analysis ; Stomach Neoplasms ; diagnosis ; metabolism ; Survival Rate

Objective To study the differential diagnostic efficacy of α-enolase ( ENO1 ) autoantibodies in lung adenocarcinoma , benign pulmonary disease , and normal individuals , and to evaluate the improvement of the diagnostic efficiency of existing markers by establishing a binary logistic regression model.Methods This was a case-control study.Participants were from the public health welfare program led by the National Cancer Center/Chinese Academy of Medical Sciences Peking Union Medical College Cancer Hospital.Serum samples were collected June 2014 to June 2017 including 60 patients with lung adenocarcinoma , 50 patients with benign lung diseases , and 90 healthy controls.Luminex MAGPIX platform was applied to detect serum ENO1 autoantibodies, CEA and Cyfra21-1 proteins.The receiver operating characteristic curve (ROC)analysis and binary logistic regression were used to evaluate the performance and build diagnostic model.Results The median level of serum ENO1 autoantibody in patients with lung adenocarcinoma was 918.5 ( 665.5-2 043.3 ), which was significantly higher than that in the normal individuals (722.5, 585.5-921.8, Z＝-3.113, P＝0.002) and benign lung disease patients (693.0, 501.4-973.3, Z＝-3.395, P＝0.001).And no significant differences between benign disease groups and normal individuals (Z＝-1.155, P＝0.248).ROC was plotted, and the area under the curve (AUC) of ENO1 autoantibodies was 0.664 (95% confidence interval : 0.576-0.752), while the AUCs of existing diagnostic marker CEA and Cyfra21-1 were 0.680 (95% confidence interval : 0.594-0.767) and 0.617 (95% confidence interval: 0.532-0.703).A joint diagnostic model including ENO1 and CEA was built with an AUC of 0.757 (95%confidence interval : 0.675-0.838).The diagnostic efficacy of the model was significantly different from ENO1 autoantibodies (Z＝2.648, P＝0.008).When the specificity was 90%, the sensitivity of ENO1 autoantibodies was 38.3%, while the sensitivity of the combination with CEA was raised to 50%.Conclusion ENO1 autoantibodies could be a marker for the auxiliary diagnosis of lung adenocarcinoma, and can improve the efficacy of the existing diagnostic markers such as CEA .ENO1 has the potential use for the diagnosis and screening.