Determination of lipemia acceptance thresholds for 31 immunoassay analytes.

BACKGROUND: Lipemia is one of common endogenous interferences that can compromises sample quality and potentially influence results of various laboratory methods. Determination of the lipemic index or triglyceride concentrations are used to define the degree of lipemia. This study was aimed to establish lipemic index (LI) and triglyceride thresholds above where significant interference exists for 31 immunoassay analytes measured on Roche Cobas 6000. MATERIALS AND METHODS: The study was carried out following CLSI C56-A and EP07-ED3:2018 guidelines using sample pools spiked with increasing concentrations of lipid emulsion solution, reaching 70 mmol/L. To define the LI and triglyceride thresholds, the bias from concentration in the native sample was calculated at different lipemia degree and compared with allowable error limits based on biological variation or state-of-the-art technology. RESULTS: No lipemia interference was observed for 27 out of 31 analytes even at the highest concentrations of lipid emulsion (LI ranging from 1737 to 2086 mg/dL, triglyceride concentration 60.34-73.99 mmol/L). However, progesterone, 25-OH vitamin D, testosterone, and estradiol were negatively affected by lipemia at 217 mg/dL (9.58 mmol/L), 222 mg/dL (10.66 mmol/L), 478 mg/dL (18.81 mmol/L), and 941 mg/dL (35.82 mmol/L) of the LI (triglyceride concentration), respectively. CONCLUSION: Most immunoassays evaluated in this study were found to be robust to lipemia interference. By using these thresholds, laboratories can report the immunoassay results from analyzing a lipemic patient sample in many cases.

Comparison of Sigma metrics computed by three bias estimation approaches for 33 chemistry and 26 immunoassay analytes.

Objectives: Sigma metric can be calculated using a simple equation. However, there are multiple sources for the elements in the equation that may produce different Sigma values. This study aimed to investigate the importance of different bias estimation approaches for Sigma metric calculation. Methods: Sigma metrics were computed for 33 chemistry and 26 immunoassay analytes on the Roche Cobas 6000 analyzer. Bias was estimated by three approaches: (1) averaging the monthly bias values obtained from the external quality assurance (EQA) studies; (2) calculating the bias values from the regression equation derived from the EQA data; and (3) averaging the monthly bias values from the internal quality control (IQC) events. Sigma metrics were separately calculated for the two levels of the IQC samples using three bias estimation approaches. The resulting Sigma values were classified into five categories considering Westgard Sigma Rules as ≥6, <6 and ≥5, <5 and ≥4, <4 and ≥3, and <3. Results: When classifying Sigma metrics estimated by three bias estimation approaches for each assay, 16 chemistry assays at the IQC level 1 and 2 were observed to fall into different Sigma categories under at least one bias estimation approach. Similarly, for 12 immunoassays at the IQC level 1 and 2, Sigma category was different depending on bias estimation approach. Conclusions: Sigma metrics may differ depending on bias estimation approaches. This should be considered when using Six Sigma for assessing analytical performance or scheduling the IQC events.

Effects of specimen haemolysis on complete blood count results by Abbott Alinity hq System.

INTRODUCTION: The current study aimed to assess the interference of in vitro haemolysis on complete blood count (CBC) using Abbott Alinity hq system, and to determine which haemolysis levels affect the reliability of sample results. MATERIALS AND METHODS: Blood samples obtained from 25 volunteers in K3-EDTA tubes were divided into four aliquots. The first aliquot was not subjected to any intervention. The second, third and fourth aliquots were passed through a fine needle 2, 4 and 6 times, respectively. Complete blood count was performed by multi-angle polarized scatter separation technology and haemolysis index (HI) was assessed from the plasma samples separated by centrifugation. Five groups were formed according to the HI values. The percentage biases between the results of non-haemolysed and haemolysed groups were compared with the desirable bias limits from The European Federation of Clinical Chemistry and Laboratory Medicine database and reference change values (RCVs). RESULTS: In groups 1 to 4, the effects of haemolysis on CBC parameters were acceptable comparing to the analytical bias except for lymphocytes (7.26%-7.42%), MCH (2.59%), and MCHC (0.47%-2.81%). Results of group 5 (gross haemolysis) showed decreases in HCT(- 4.56%), RBC (- 4.07%) count and increase in lymphocyte (11.60%) count higher than the analytical performance specifications. Moreover, variations in MCH (4.65%) and MCHC (5.24%) were exceeding the RCVs. CONCLUSIONS: Gross haemolysis (haemoglobin concentration > 10 g/L) is likely to produce unreliable CBC results on non-pathological samples. Further studies including pathological specimens are needed.

Establishing biological variation for plasma D-dimer from 25 healthy individuals.

D-dimer is considered to be a reliable marker of both coagulation activation and fibrinolysis. However, data on biological variation (BV) of D-dimer is still limited, causing the use of empiric analytical performance specifications and lack of other implications related to BV. This study aimed to estimate the BV of plasma D-dimer employing a study design compliant with The Biological Variation Data Critical Appraisal Checklist. Blood samples were collected from a cohort of 25 healthy subjects (16 females, 9 males; age range, 19-61 years) from Turkey once weekly for 3 consecutive weeks. All plasma samples were analyzed in duplicate within a single run on Roche Cobas c501. The results were assessed for outliers, variance homogeneity, normal distribution, and trend, followed by nested ANOVA to determine BV and analytical variation estimates with confidence intervals (CIs). Gender stratified BV estimates were also calculated. Within-subject (CVI) and between-subject (CVG) BV estimates with 95% CIs were for D-dimer 21.2% (17.8-25.9) and 30.9% (21.3-46.2), respectively. No significant BV differences were observed between females and males. The index of individuality (II) and the reference change value (RCV) were calculated as 0.71 and 60.4%, respectively. Analytical performance specifications for desirable imprecision, bias, and total error were 10.6, 9.4, and 26.8%, respectively. This study provides well-characterized BV estimates for D-dimer, which may be helpful for setting objectively analytical performance specifications. Moreover, RCV should be preferred to decide whether a significant difference is present between serial D-dimer measurements from an individual.

Order of draw of blood samples affect potassium results without K-EDTA contamination during routine workflow.

INTRODUCTION: A specific sequence is recommended for filling blood tubes during blood collection to prevent erroneous test results due to carryover of additives. However, requirement of this procedure is still debatable. This study was aimed to investigate the potassium ethylenediaminetetraacetic acid (K-EDTA) contamination in blood samples taken after a tube containing the additive during routine workflow. The study was also carried out to examine the effect of order of draw on potassium results, regardless of K-EDTA contamination. MATERIALS AND METHODS: In 388 outpatients, to determine the probability of K-EDTA cross-contamination, blood was drawn sequentially into a serum tube, followed by a tube containing K-EDTA, and by another serum tube. In another 405 outpatients, to evaluate the effect of order of draw blood unrelated to K-EDTA contamination, two serum tube were successively collected. Potassium was measured on Cobas 6000 c501 analyser (Roche Diagnostic GmbH, Mannheim, Germany) by indirect ion selective electrode method. RESULTS: Of paired samples collected before and after a K-EDTA tube, 24% had a potassium difference of above 0.3 mmol/L. However, no EDTA contamination was detected in these samples as well as 95% confidence intervals (CI) of limits of agreement for calcium were within the allowable error limits based on reference change values. Interestingly, of blood samples drawn successively, 24% had also a difference greater than 0.3 mmol/L for potassium. CONCLUSION: Incorrect order of draw using closed blood collection system does not cause K-EDTA contamination, even in routine workflow. However, regardless of K-EDTA contamination, order of draw has significant influence on the potassium results.

The accuracy of transcutaneous bilirubinometer measurements to identify the hyperbilirubinemia in outpatient newborn population.

OBJECTIVES: To assess the ability of transcutaneous bilirubinometer for prediction total serum bilirubin levels in newborn infants after hospital discharge. METHODS: Newborn infants requiring total serum bilirubin (TSB) level measurement during an outpatient follow-up visit were included into the study. Transcutaneous bilirubin (TcB) measurement was carried out using JH20-1C (Ningbo David, China) transcutaneous jaundice detector and total serum bilirubin was simultaneously determined by direct spectrophotometry. The agreement between paired TSB and TcB measurements were assessed by Bland-Altman plot and Deming regression analysis. Predictive indices were also identified in different TcB cut-off values for TSB levels of 222, 256 and 291 µmol/L. RESULTS: A total of 271 paired TcB and TSB measurements were obtained from 218 newborn infants. 40.2% had an age of above 7 days at measurement time. The mean difference (95% CI) between TcB and TSB values was -1.7 (-5.4 to 2.1) µmol/L. For TSB levels of at least 256 and 291 µmol/L, a TcB cut-off of 222 µmol/L shows sensitivity of 90.6% and 100%, respectively. It was also determined that 39.4% of TSB measurements could be avoided when using TcB cut-off value of 222 µmol/L. CONCLUSION: The measurement of TcB with JH20-1C seems to be a reliable screening method for hyperbilirubinemia in the outpatient population when used a TcB cut-off of 222 µmol/L. The use of transcutaneous bilirubinometer could reduce the number of invasive blood sampling for the determination serum bilirubin.

Does preeclampsia have any adverse effect on fetal heart?

OBJECTIVE: To determine whether preeclampsia causes fetal cardiac cell damage by assessing umbilical artery NT-proBNP, cardiac troponin I and homocysteine. METHODS: A cross-sectional study with 73 fetuses between 26 and 40 weeks of gestation was performed. Thirty-three healthy mothers' fetuses were control group (Group I). While 12 mildly pre-eclamptic mothers' fetuses constituted Group II, 28 fetuses of severe pre-eclamptic mothers were Group III. RESULTS: Umbilical cord mean NT-proBNP levels of Group I, II and III are 520.8 ± 404.5 pg/ml; 664.2 ± 215.9 pg/ml; and 1932.8 ± 2979.5 pg/ml, respectively (p = 0.0001). The number of neonates with NT-proBNP > 500 pg/mL that indicates severe cardiac damage is higher in Group III (p = 0.001). The mean homocysteine levels are also statistically significantly higher in Group III. Cardiac troponin I levels are not different between the groups (p = 0.46). CONCLUSION: Increased NT-proBNP and homocysteine might not only indicate some degree of in-utero cardiac cell damage but also feto-placental endothelial injury in the fetuses of severe pre-eclamptic mothers. Our finding that shows no evidence of correlation between cardiac troponin I levels with cell damage and endothelial injury requires further research.

The relationship between serum adiponectin and resistin levels, insulin resistance and colorectal adenomas.

BACKGROUND/AIMS: The relationship between adipocytokines and the development of colorectal cancer is well-documented. Our aim was to assess the relationship among serum adiponectin and resistin levels, insulin resistance, and colorectal adenoma to evaluate whether these parameters can be used as biomarkers to predict the development of colorectal adenoma. MATERIALS AND METHODS: This is a cross-sectional case-control study conducted in 32 patients with colorectal adenoma and 30 control subjects. Serum adiponectin and resistin levels, body mass index values, waist and hip circumferences and Homeostasis Model Assessment scores were measured. RESULTS: Resistin levels were slightly higher and adiponectin was slightly lower in patients with colorectal adenoma compared with controls; however, the differences in both parameters failed to reach statistical significance. The body mass index values and waist circumference of the patient group were significantly higher than controls (p=0.003 and p=0.002, respectively). Fasting serum insulin levels and Homeostasis Model Assessment scores of patients with colorectal adenoma were significantly higher than those of controls (p=0.02 and p=0.02, respectively). There was no relation between the number of colorectal adenomas and serum adiponectin or resistin levels. CONCLUSION: Our data indicate that obesity and insulin resistance may contribute to the development of colorectal adenoma and that serum adiponectin levels and insulin resistance may not have a substantial predictive value for colorectal adenoma.

Establishing reference intervals for D-dimer to trimesters.

OBJECTIVE: When D-dimer is used to evaluate suspected venous thromboembolism in pregnant patients, the reference interval of common population may cause misinterpretation. The present study aims to determine reference intervals of D-dimer in the three trimesters. METHODS: Four-hundred sixteen pregnant women and 32 non-pregnant women were enrolled in this cross-sectional study. Reference group had comprised 123 pregnant in the first trimester (5-11 week), 164 pregnant in the second trimester (13-20 week) and 126 pregnant women in the third trimester (25-35 week). D-dimer levels were analyzed via immunoturbidimetric assay. RESULTS: If the threshold of 0.50 mg/L for diagnosis of VTE is used, 4.8% of pregnant women in the second trimester and 23.8% of pregnant women in the third trimester would have D-dimer levels exceeding this cut-off value. Reference intervals of D-dimer were determined as 0.11-0.40 mg/L; 0.14-0.75 mg/L and 0.16-1.3 mg/L in first, second and third trimester, respectively. CONCLUSION: The established D-dimer reference intervals for each trimester of pregnancy are different from those used in common population. These reference intervals may assist clinicians in making accurate clinical decisions. Further studies are needed to establish new cut-off values for the D-dimer to rule out VTE in each trimester.

The changing of serum vitamin B12 and homocysteine levels after gastrectomy in patients with gastric cancer: do they associate with clinicopathological factors?

After total (TG) or distal subtotal gastrectomy (DG), patients are at high risk of vitamin B12 (vit-B12) deficiency, which results in elevation of homocysteine levels. The changing of serum vit-B12 and homocysteine levels in patients with gastric cancer is not well known. Seventy-two patients with gastric cancer who had undergone currative gastrectomy and 50 healthy controls were included. Serum vit-B12 and homocysteine levels were analyzed in gastric cancer patients. In addition, these parameters were compared with those of healthy control subjects. While serum vit-B12 levels in gastrectomized patients were significantly lower than that of healthy controls (221.8 ± 125.6 pg/mL vs. 309.9 ± 174.3 pg/mL, p = 0.002), homocysteine levels were significantly higher in patients with gastric cancer (14.2 ± 6.7 µmol/L vs. 12.5 ± 6.1 µmol/L, p = 0.016). Mean serum folate level was found to be high in healthy controls (7.3 ng/mL) compared to patients (9.2 ng/mL, p = 0.027). Out of 72 patients, 40 patients (55.6 %) with gastric cancer developed vit-B12 deficiency after gastrectomy. Vit-B12 deficiency was found to be related with gastrectomy type (p = 0.02) and homocysteine levels (p = 0.014). In patients who underwent TG, the incidence of vit-B12 deficiency was significantly higher compared with those with DG (67.5 vs. 32.5 %). In addition, serum vit-B12 level in patients with DG was significantly higher than that of patients with TG (248.3 ± 122.0 pg/mL vs. 200.8 ± 126.7 pg/mL, p = 0.041), whereas homocysteine levels were significantly lower in DG group compared with TG group (12.1 ± 6.1 µmol/L vs. 15.8 ± 6.9 µmol/L, p = 0.014). A logistic regression analysis showed that the extent of gastrectomy was found to be an independent factor for predicting the occurrence of vit-B12 deficiency (p < 0.001, odds ratio 1.38). Our results showed that cumulative vit-B12 deficiency rate was significantly higher after TG compared with that after DG, while homocysteine levels were significantly higher in TG group compared with DG group. The extent of gastrectomy was found to be an independent factor for predicting the occurrence of vit-B12 deficiency. Vit-B12 deficiency and hyperhomocysteinemia are imperious clinical situation for patients with gastric cancer after surgery. Hence, both preoperative and regular postoperative monitoring of vit-B12 and homocysteine levels for all gastrectomized patients with gastric cancer are important and necessary for early detection and prevention of vit-B12 deficiency and hyperhomocysteinemia as a risk factor for cardiovascular diseases.

70-year old female patient with mismatch between hematocrit and hemoglobin values: the effects of cold agglutinin on complete blood count.

INTRODUCTION: There are a number of pre-analytical and analytical factors, which cause false results in the complete blood count. The present case identifies cold agglutinins as the cause for the mismatch between hematocrit and hemoglobin values. MATERIALS AND METHODS: 70-year old female patient had a history of cerebrovascular diseases and rheumatoid arthritis. During routine laboratory examination, the patient had normal leukocyte and platelet counts; however, the hemoglobin (Hb: 105 g/L) and hematocrit (HCT: 0.214 L/L) results were discordant. Hemolysis, lipemia and cold agglutinin were evaluated as possible reasons for the mismatch between hematocrit and hemoglobin values. RESULTS: First blood sample was slightly hemolysed. Redrawn sample without hemolysis or lipemia was analyzed but the mismatch became even more distinct (Hb: 104 g/L and HCT: 0.08 L/L). In this sample, the titration of the cold agglutinin was determined and found to be positive at 1:64 dilution ratios. After an incubation of the sample at 37°C for 2 hours, reversibility of agglutination was observed. CONCLUSION: We conclude that cold agglutinins may interfere with the analysis of erythrocyte and erythrocyte-related parameters (HCT, MCV, MCH and MCHC); however, Hb, leukocyte and platelet counts are not affected.

The prognostic importance of changing serum M30 and M65 values after chemotherapy in patients with advanced-stage non-small-cell lung cancer.

Although oncological treatments are improving, the prognosis of non-small-cell lung cancer (NSCLC) patients has not. Several biomarkers related to prognosis have been evaluated, and M30 and M65 have been reported to be higher in patients with NSCLC than in healthy people. In the current study, we evaluated the clinical importance of the change in serum M30 and M65 values after chemotherapy in patients with NSCLC. Serum M30 and M65 values were measured before and 48 h after chemotherapy in thirty-two patients with advanced NSCLC. The importance of the change in the levels of these markers after chemotherapy was analyzed by univariate analysis. The median serum M65 and M30 values increased significantly after chemotherapy (p < 0.001). The median M30 value after chemotherapy was an important prognostic factor for both overall survival (OS) (p = 0.002) and progression-free survival (PFS) (p = 0.002). Stage and histopathological type were significant both for PFS and OS. Multivariate analysis showed that the median M30 value after chemotherapy was the only independent prognostic factor for PFS (p = 0.04, HR 5.4) and OS (p = 0.02, HR 11.49). Our results indicated that both serum M30 and M65 values increased after chemotherapy in patients with advanced NSCLC, and an elevated serum M30 value was an independent prognostic factor for both PFS and OS.

The prognostic significance of the increase in the serum M30 and M65 values after chemotherapy and relationship between these values and clinicopathological factors in patients with advanced gastric cancer.

In some studies, the prognostic and predictive significance of M30 and M65 has been reported to detect response to chemotherapy. In the present study, we aimed at determining the changes of serum M30 and M65 values after chemotherapy and the impact of these values on treatment response and progression-free survival (PFS) and overall survival (OS) of patients with advanced gastric cancer. A total of 31 patients with advanced gastric cancer was included. M30 and M65 values were measured by a quantitative enzyme-linked immunosorbent assay (ELISA) method in serum samples before and 48 h after the first chemotherapy cycle. Pre- and postchemotherapy values of M30 and M65 were compared. The difference between the mean values of serum M30 and M65 before and after chemotherapy was calculated and the prognostic significance of changes for survival was evaluated by univariate and multivariate analysis. Logistic regression analysis was performed to predict response to chemotherapy. Serum M30 and M65 levels were found to be increased significantly after chemotherapy (M30, 582.7 ± 111.5 U/l [pre mean] vs. 983.3 ± 214.1 U/l [post mean], p = 0.01; M65, 2,061.7 ± 431.2 U/l [pre mean] vs. 2,646.3 ± 433.1 U/l [post mean], p = 0.003). Means of the differences of M30 and M65 levels before and 48 h after chemotherapy were 400.5 ± 190 U/l ([M30-difference] M30-D) and 584.6 ± 335.4 U/l (M65-D), respectively. Patients with serum M30-D of <400.5 U/l had better median PFS and OS times than patients with M30-D >400.5 U/l (PFS, 9.9 vs. 4.3 months, p = 0.018 and OS, 13.6 vs. 8.1 months, p = 0.029). In addition, median PFS and OS intervals in patients with serum M65-D > 584.6 U/l were significantly worse than those of patients whose M65-D was lower than or equal to 584.6 U/l (4.1 vs. 11.4 months for PFS, p = 0.002 and 5.7 vs. 13.6 months for OS, p = 0.005). Patients with values above M30-D and M65-D had a better tumor response compared with patients with values below M30-D and M65-D (p = 0.02 and p = 0.006, respectively). In the logistic regression analysis, only M65-D was significantly found to be an independent factor in predicting response to chemotherapy (p = 0.018, OR:1.4). However, only M30 levels after chemotherapy were found to be an independent prognostic factor for PFS in the multivariate analysis. These results showed for the first time that both M30 and M65 in serum samples of patients with advanced gastric cancer were elevated 48 h after chemotherapy and these were poor prognostic factors for both PFS and OS of patients. Moreover, increased serum M65 levels after chemotherapy can be predict tumor response.

Serum M30 and M65 values in patients with advanced stage non-small-cell lung cancer compared with controls.

BACKGROUND: M30 and M65 are derivatives of cytokeratin 18 and released from the epithelial cell during cell death. These markers can be used to evaluate prognosis and chemotherapy response in several tumours. We evaluated serum M30 and M65 values in patients with advanced nonsmall- cell lung cancer (NSCLC) compared with those in a healthy group. MATERIAL AND METHODS: Thirty-two patients with advanced NSCLC and thirty-two healthy people were included in the study. Serum M30 and M65 values were measured by quantitative ELISA method. The best cut-off value for serum M65 was calculated by ROC analysis and then univariate analysis was performed to determine the importance of M65 value in predicting progression-free survival (PFS). RESULTS: There were no differences between mean serum M30 values between patients and controls (445.44±536.17 vs. 340.56±345.07, p=1). The mean serum M65 values were found to be significantly higher in patients than in healthy controls (1421.30±1662.59 vs. 648.85±341.17, p<0.001). The best cut-off value for serum M65 predicting PFS was 1311.64 U/l (AUC 0.58, sensitivity and specificity were 45.5% and 85.7% respectively). The patients with serum M65 values ≥1311.64 U/l had worse PFS than patients with serum M65 values <1311.64 U/l, p=0.01). There was no correlation between serum M30 value and PFS in the patient group (p=0.4). CONCLUSIONS: Our results indicated that serum M65 values elevated in advanced NSCLC compared to a healthy control group and elevated serum M65 level can predict PFS in patients.

Serum concentrations of CA 125, CA 15-3, CA 19-9 and CEA in normal pregnancy: a longitudinal study.

PURPOSE: Although cancer diagnosed during pregnancy is rare, the coexistence of pregnancy and malignancy becomes more common in view of prolongation of reproductive age. Therefore, it is important that the specificity of a tumor marker be evaluated during pregnancy to avoid misinterpretation in the follow-up of a pregnant cancer patient. The present study aims to investigate the serum concentrations of CA-125, CA 15-3, CA 19-9 and CEA in healthy pregnant women through gestation. METHODS: In this prospective study, we followed thirty healthy pregnant women. Blood samples were obtained during each trimester of pregnancy (10-12, 22-24 and 34-36 weeks). The maternal serum levels of CA-125, CA 15-3, CA 19-9 and CEA were measured using electrochemiluminescence immunoassay. RESULTS: There was no difference between the first and second trimester serum levels of CA 125, CEA and CA 19-9. However, serum CA 125 levels in third trimester were found to be significantly elevated in pregnants compared to the second trimester (median values 19.6 vs. 15.6 IU/mL, p = 0,009). Similarly, the serum CEA levels in third trimester were significantly higher than those of second trimester (median values 1.1 vs. 0.7 ng/ml, p = 0.001). It is also found that CEA and CA 19-9 assay values were significantly elevated in the third trimester of pregnancy when compared with the first trimester of pregnancy (CEA median values 1.1 vs. 0.7 ng/ml, p = 0.02 and CA 19-9 median values 11.6 vs. 7.7 IU/mL, p = 0,02). Three trimester had statistically similar levels for serum CA 15-3 (median values 17.5, 19.7 and 18.3 U/mL, respectively). The four tumor markers assay values were found generally within the normal range. CONCLUSIONS: These findings suggest that maternal serum levels of CA 125, CEA and CA 19-9 were increased during third trimester of pregnancy. However, these elevations were within the normal range. CA 15-3 is independent of gestation and reliable tumor markers in monitoring malignancy in pregnant patients.

Is there any impact of plasma M30 and M65 levels on progression-free survival of patients with advanced gastric cancer?

PURPOSE: M30 and M65 are different circulating fragments of cytokeratin 18. They release during apoptotic cell death, so it is believed that they reflect cell death of epithelial tumors. The aim of this study was to determine the prognostic value of plasma M30 and M65 levels in predicting of survival for patients with advanced gastric cancer compare with healthy controls. METHODS: Thirty-four patients with advanced gastric cancer and thirty-two healthy controls were included. Plasma M30 and M65 values were measured by quantitative ELISA method. RESULTS: The median age of patients and control groups was 60 and 56 years, respectively. No difference was detected between patient and control groups with respect to plasma median M30 values (390.4 vs. 270.7 U/l, respectively, P = 0.10). The median plasma M65 values of patients were significantly higher than those of control group (1232.1 vs. 580.1 U/l, P < 0.001). The best cut-off values for plasma M30 and M65 for predicting progression-free survival (PFS) were 277.7 and 1434.9 U/l in ROC analysis. The patients whose plasma M30 values were higher than 277.7 U/l had worse PFS than patients with plasma M30 value <277.7 U/l (8.9 vs. 11.2, respectively, P = 0.01). The median PFS of patients whose M65 levels lower than or equal to 1434.9 U/l was better than that of patients whose M65 levels were >1434.9 U/l (12.4 vs. 10.4, respectively, P = 0.04). But plasma M30 and M65 level in patient group were not found to be an important prognostic factor for PFS in the multivariate analysis. CONCLUSIONS: These results showed that plasma M65 values were significantly elevated in patients with advanced gastric cancer compared to healthy people. Moreover, both increased plasma M30 and M65 levels can predict PFS in patients with gastric cancer.