1、Optimization of TAT antibody working concentration
The experiment utilized the checkerboard method with biotin-labeled and HRP-labeled antibodies. We selected various concentrations of antibodies for detection and determined the relative light units (RLUs) of six standards. The results are presented in Table 1. The working concentration of the antibody for this study was chosen by selecting a lower antibody concentration under the same conditions. This decision was made after confirming that the signal value of each calibrator was appropriate and the ratio was close when the concentration of TAT-Biotin was 1 μg/mL and the concentration of TAT-HRP was 1.5 μg/mL. We are confident that this combination will effectively reduce costs without compromising the quality of the study.
Table 1 TAT Determining the concentration of Biotin-labeled antibody with TAT HRP-labeled antibody
TAT Biotin-labeled antibody (μg/mL)
|
0.5
|
1
|
1.5
|
TAT HRP-labeled antibody (μg/mL)
|
1
|
1.5
|
2
|
1
|
1.5
|
2
|
1
|
1.5
|
2
|
S1
|
3045
|
4967
|
44457
|
2987
|
5823
|
6745
|
4981
|
5019
|
7851
|
S2
|
39671
|
45781
|
56871
|
40086
|
82997
|
78751
|
89761
|
69812
|
80127
|
S3
|
233879
|
346781
|
398721
|
400763
|
404211
|
569021
|
427638
|
600927
|
676528
|
S4
|
908745
|
1009756
|
1276545
|
1198782
|
1693406
|
1789657
|
1839007
|
2009813
|
2319707
|
S5
|
3987876
|
4657825
|
5086372
|
5902638
|
6529542
|
6908263
|
5278297
|
6093821
|
6273849
|
S2/S3
|
17%
|
13%
|
14%
|
10%
|
21%
|
14%
|
21%
|
12%
|
12%
|
S3/S4
|
26%
|
34%
|
31%
|
33%
|
24%
|
32%
|
23%
|
30%
|
29%
|
S4/S5
|
23%
|
22%
|
25%
|
20%
|
26%
|
26%
|
35%
|
33%
|
37%
|
2、Optimal incubation time
The length of incubation time affects the RLU, and different times were chosen to test the standards 1-5, low and high value plasma samples, and the results are shown in Figure 1 and Table 2. At 10 min of incubation, the signal value of each calibrator was higher overall, and selecting a shorter incubation time can significantly reduce the whole test time and improve the detection throughput, so the reaction conditions of this kit were set to 37°C for 10 min.
Table 2 TAT kit 37 °C reaction time determination experiment
reaction time
|
5 min
|
10 min
|
20 min
|
30 min
|
standards 1
|
4903
|
5832
|
4981
|
3301
|
standards 2
|
70993
|
82997
|
72897
|
63782
|
standards 3
|
337208
|
414211
|
362036
|
298730
|
standards 4
|
1128936
|
1893406
|
1409729
|
1029837
|
standards 5
|
5039483
|
6529542
|
5698345
|
4128398
|
3. Method performance evaluation
Method performance was evaluated according to the Clinical and Laboratory Standards Institute (CLSI) EP guidelines.
3.1 Limit of detection
The mean and standard deviation of RLU of 20 standard dilutions were calculated, and M+2SD was brought into the standard curve to obtain the concentration values of 0.045 ng/mL, 0.048 ng/mL, 0.039 ng/mL, respectively, so the lowest detection limit of the method was 0.048 ng/mL, which is shown in Table 3.
Table 3 Verification results of the limit of detection
Indicators
|
Test1
|
Test2
|
Test3
|
M
|
5459.8
|
5460.8
|
5461.8
|
SD
|
338.58
|
289.72
|
312.35
|
M+2SD
|
6164.16
|
6265.44
|
5978.70
|
LOD (ng/mL)
|
0.045
|
0.048
|
0.039
|
Abbreviation: M, Mean; SD, Standard deviation; LOD, Limit of detection.
3.2 Linearity
The high-value samples close to the upper limit of the linear range were diluted into a number of concentrations in a certain proportion, of which the low-value concentration samples had to be close to the lower limit of the linear range. The results of linear fitting by least squares method with the calibration product concentration as the horizontal coordinate and the mean value of RLU as the vertical coordinate (Figure 2) are as follows, which shows that there is a good linear relationship between the theoretical concentration of the samples and the actual concentration, and the linear regression equation after the fitting is y=0.9914x-0.6358, R=0.9995.
3.3 Accuracy
The assay was repeated three times for the high-value QC and low-value QC, respectively, and the average value of the measured concentration results was recorded as (M), and the relative deviation (Bi) was calculated according to Bi(%)=(M-T)/T*100%, respectively, where Bi is the relative deviation, M is the average value of the measured concentration, and T is the theoretical concentration, and the results were shown in Table 4, and the relative deviations were all within 5%.
Table 4 TAT accuracy measurement results
Concentrations (ng/mL)
|
Test1
|
Test2
|
Test3
|
Average (ng/mL)
|
Bias (%)
|
1.3
|
1.365
|
1.333
|
1.391
|
1.363
|
4.62%
|
26.0
|
27.213
|
26.585
|
27.684
|
27.161
|
4.46%
|
3.4 Precision
Three batches of reagents were prepared, and the test was repeated 10 times for the high-value QC and low-value QC, respectively, and the mean (M) and standard deviation (SD) of the results were calculated, and the coefficient of variation (CV) was calculated according to CV (%)=SD/M*100%, where CV is the coefficient of variation, M is the mean of the tested concentration, and SD is the standard deviation, and the results were shown in Table 5, with the precision within 3%.
Table 5 Comparison of TAT precision measurement results
The name of the project
|
TAT assay results
|
Lot 1
|
Lot 2
|
Lot 3
|
Concentration level
|
10 ng/mL
|
40 ng/mL
|
10 ng/mL
|
40 ng/mL
|
10 ng/mL
|
40 ng/mL
|
1
|
11.52
|
46.12
|
11.07
|
47.03
|
11.22
|
47.24
|
2
|
11.84
|
46.44
|
11.38
|
46.44
|
11.55
|
47.12
|
3
|
11.53
|
47.38
|
11.47
|
47.38
|
11.75
|
47.38
|
4
|
11.72
|
47.02
|
11.55
|
46.02
|
11.63
|
47.02
|
5
|
11.27
|
48.52
|
11.35
|
47.52
|
11.88
|
47.52
|
6
|
11.41
|
48.65
|
11.22
|
48.65
|
11.51
|
48.65
|
7
|
12.11
|
49.32
|
11.41
|
49.32
|
11.99
|
49.32
|
8
|
11.55
|
49.82
|
11.27
|
48.50
|
12.05
|
50.81
|
9
|
12.07
|
48.95
|
11.15
|
49.23
|
11.97
|
49.31
|
10
|
11.83
|
48.77
|
11.95
|
49.88
|
12.03
|
50.71
|
Average
|
11.685
|
48.099
|
11.382
|
47.997
|
11.758
|
48.508
|
SD
|
0.28
|
1.27
|
0.25
|
1.31
|
0.28
|
1.47
|
CV%
|
2.37%
|
2.64%
|
2.17%
|
2.72%
|
2.34%
|
3.03%
|
3.5 Acceleration Stability
Five batches of reagents were placed in a 37℃ constant temperature incubator, and samples were taken on the 1, 4 and 7 days for testing, comparing with the reagents stored at 4℃, and analyzing the performance indexes. The results, as shown in Table 6, showed that the deviations of the kits placed at 37°C for different times compared with those stored at 4°C were less than 10%, indicating that the stability of the method was good.
Table 6 Accelerated stability test results
|
S1
|
S2
|
S3
|
S4
|
S5
|
test1
|
test2
|
test1
|
test2
|
test1
|
test2
|
test1
|
test2
|
test1
|
test2
|
4°1Day
|
5823
|
5844
|
80097
|
82289
|
394211
|
403916
|
1804187
|
1823406
|
6418172
|
6429542
|
37°1Day
|
5733
|
5667
|
78976
|
78758
|
388867
|
388245
|
1784274
|
1733298
|
6112683
|
6071570
|
37°4Day
|
5625
|
5634
|
79749
|
78411
|
384194
|
385240
|
1775541
|
1792015
|
6256376
|
6103575
|
37°7Day
|
5559
|
5588
|
77185
|
77160
|
375501
|
376844
|
1692983
|
1752580
|
6021054
|
6060396
|
Day 1 decline
|
-2%
|
-3%
|
-1%
|
-4%
|
-1%
|
-4%
|
-1%
|
-5%
|
-5%
|
-6%
|
Day 2 decline
|
-3%
|
-4%
|
0%
|
-5%
|
-3%
|
-5%
|
-2%
|
-2%
|
-3%
|
-5%
|
Day 3 decline
|
-5%
|
-4%
|
-4%
|
-6%
|
-5%
|
-7%
|
-6%
|
-4%
|
-6%
|
-6%
|
3.6 Clinical evaluation
The HISCL-800 kit from Sysmex (Japan) was used to test 130 clinical plasma samples. The analysis resulted in a correlation coefficient of R2 = 0.9683, indicating excellent agreement between the two assays (Figure 3).
4、General features
A total of 177 patients with cervical cancer were included, including 42 cases of stage I, 45 cases of stage II, 62 cases of stage III and 28 cases of stage IV. The mean age of the thrombosis group was 52.37±11.35 years, and the mean age of the no thrombosis group was 55.27±10.24 years. Other general information is shown in Table 7.
Table 7 General information of cervical cancer patients
Characteristics
|
|
All (n=177)
|
Mean ± SD or no. (%)
|
Age (year)
|
|
54.83±10.437
|
Menopausal status
|
Pre-menopausal
|
82 (46.328%)
|
Post-menopausal
|
95 (53.672%)
|
Pathologic stage
|
Ⅰ
|
42 (23.729%)
|
Ⅱ
|
45 (25.424%)
|
Ⅲ
|
62 (35.028%)
|
Ⅳ
|
28 (15.819%)
|
Blood clots occur
|
Yes
|
27 (15.254%)
|
No
|
150 (84.746%)
|
Traditional thrombosis testing
|
D-Dimer
|
1.472±2.531
|
PT
|
11.131±0.886
|
APTT
|
26.737±3.285
|
FDP
|
4.997±7.267
|
Four tests for thrombosis
|
TAT
|
12.414±9.173
|
t-PAIC
|
6.222±3.689
|
PIC
|
0.822±1.146
|
TM
|
6.783±4.572
|
Treatment
|
Surgery
|
61 (34.463%)
|
Surgery and chemotherapy
|
42 (23.723%)
|
Surgery and radiotherapy
|
3 (1.694%)
|
Surgery and chemotherapy and radiotherapy
|
18 (10.169%)
|
Chemotherapy and(or) radiotherapy
|
53 (29.943%)
|
5、Comparison of t-PAIC, TAT, TM, and PIC between healthy control and cervical cancer groups
As shown in Table 8, we found that TAT and PIC were significantly higher than those of the control group, and the difference was statistically significant (P < 0.001 or P < 0.05).
Table 8 Comparison of thrombotic markers between healthy individuals and cervical cancer patients
Thrombosis markers
|
Control Group (n=80)
|
Cervical cancer group (n=177)
|
Z Value
|
P Value
|
t-PAIC (ng/mL)
|
4.49 (3.04-7.21)
|
5.82 (4.07-7.94)
|
-2.508
|
0.012
|
TAT (ng/mL)
|
3.27 (2.65-3.91)
|
10.43 (6.61-13.87)
|
-11.251
|
<0.001
|
TM (TU/mL)
|
6.27 (4.49-8.44)
|
5.57 (4.02-10.92)
|
-0.177
|
0.860
|
PIC (ng/mL)
|
0.32 (0.20-0.50)
|
0.70 (0.44-0.95)
|
-7.839
|
<0.001
|
Abbreviation: VTE: Stomach cancer; t-PAIC: Tissue plasminogen activator inhibitor complex; TAT: Thrombin-antithrombin complex; TM: Thrombomodulin; PIC:α2-plamininhibitor-plasmin complex.
*P < 0.05 considered statistically significant.
6、Comparison of t-PAIC, TAT, PIC, TM, D-D, and FDP between thrombus and no thrombus groups of cervical cancer patients
As shown in Table 9, TAT and FDP in the thrombus group were significantly higher than those in the non-thrombus group, and the difference was statistically significant (P < 0.001, t-PAIC) The difference was not statistically significant.
Table 9 Comparison of Thrombosis Markers between Two Groups of Patients
Thrombosis markers
|
Thrombosis group (n=27)
|
Non-thrombosis group (n=150)
|
Z Value
|
P Value
|
t-PAIC (ng/mL)
|
11.29 (4.10-13.72)
|
5.63 (4.04-7.27)
|
-2.811
|
0.005
|
TAT (ng/mL)
|
19.07 (13.61-31.84)
|
9.43 (6.36-12.64)
|
-4.806
|
<0.001
|
TM (TU/mL)
|
14.33 (4.14-15.44)
|
5.22 (3.96-7.93)
|
-3.117
|
0.002
|
PIC (ng/mL)
|
0.94 (0.82-1.26)
|
0.68 (0.43-0.92)
|
-3.466
|
0.001
|
FDP (mg/L)
|
6.30 (2.80-7.10)
|
2.65 (1.80-4.30)
|
-3.604
|
<0.001
|
D-D (mg/L)
|
0.70 (0.53-1.36)
|
0.40 (0.26-0.48)
|
-3.303
|
0.001
|
7、Comparison of t-PAIC, TAT, PIC, TM, D-D, and FDP in patients with stage I-II versus stage III-IV cervical cancer
As shown in Table 10, there was no significant difference between the two groups of data in terms of t-PAIC, TM, and PIC concentrations (P > 0.05); in terms of TAT concentration, there was a significant difference between the two groups of data (P < 0.05), and the TAT concentration of patients with advanced stages showed a significant trend of increase compared with that of patients with early stages.
Table 10 Comparison of thrombosis markers in patients with early and advanced cervical cancer (x±s)
Thrombosis markers
|
Stage Ⅰ-Ⅱ (n=87)
|
Stage Ⅲ-Ⅳ (n=90)
|
Z Value
|
P Value
|
t-PAIC (ng/mL)
|
5.38 (3.93~7.05)
|
6.50 (4.10~10.36)
|
-2.033
|
0.042
|
TAT (ng/mL)
|
8.77 (5.94~11.71)
|
11.32 (8.06~17.18)
|
-3.531
|
<0.001
|
TM (TU/mL)
|
4.81 (3.79~7.41)
|
7.30 (4.20~14.59)
|
-3.565
|
0.001
|
PIC (ng/mL)
|
0.65 (0.42~0.92)
|
0.80 (0.47~0.99)
|
-2.099
|
0.036
|
FDP (mg/L)
|
2.30 (1.70~4.50)
|
3.30 (2.20~5.25)
|
-2.429
|
0.015
|
D-D (mg/L)
|
0.38 (0.23~0.49)
|
0.45 (0.31~0.74)
|
-2.043
|
0.041
|
8、Evaluation of the value of each marker in the diagnosis of VTE in patients with cervical cancer using receiver operating characteristic (ROC) curve analysis
For the diagnosis of VTE, the area under the curve (AUC) of t-PAIC, TM, TAT, PIC, D-Dimer, and FDP were 0.670, 0.689, 0.791, 0.710, 0.700, and 0.718, respectively. t-PAIC, TM, TAT, PIC, D-Dimer, and FDP had critical values of 10.580 ng /mL, 11.624 TU/mL, 15.627 ng/mL, 0.822 ng/mL, 0.515 mg/L, and 5.000 mg/L, respectively. TAT had a good diagnostic value for VTE, with a sensitivity of 74.1% and a specificity of 91.3%. t-PAIC, with an AUC of 0.670, was found to have the lowest diagnostic efficiency (Table 11 and Figure 4-A). When the six markers were applied in combination, the AUC of "t-PAIC+TM+TAT+PIC" and "t-PAIC+TM+TAT+PIC+D-D+FDP" were significantly higher than the AUC of "D-D+FDP" for VTE in cervical cancer patients. "The AUC, sensitivity and specificity of "D-D+FDP" were 0.688, 51.9% and 88.7%, respectively, while that of "t-PAIC+TM+TAT+PIC" was 0.688, 51.9% and 88.7%, respectively. TAT+PIC" had an AUC, sensitivity and specificity of 0.894, 81.5% and 91.3%, respectively. The AUC, sensitivity and specificity of "t-PAIC+TM+TAT+PIC+D-D+FDP" were 0.912, 85.2% and 88.7%, respectively (Table 12, Figure 4-B).
Table 11 Receiver operating curve (ROC) analysis for each test marker
Markers
|
Cut-off values
|
AUC (95%CI)
|
Sensitivity (%)
|
Specificity (%)
|
Youden Index
|
t-PAIC
|
10.580 ng/mL
|
0.670 (0.529-0.812)
|
59.3
|
92.0
|
0.513
|
TM
|
11.624 TU/mL
|
0.689 (0.570-0.808)
|
63.0
|
83.3
|
0.463
|
TAT
|
15.627 ng/mL
|
0.791 (0.663-0.919)
|
74.1
|
91.3
|
0.654
|
PIC
|
0.822 ng/mL
|
0.710 (0.594-0.825)
|
77.8
|
69.3
|
0.471
|
D-Dimer
|
0.515 mg/L
|
0.700 (0.590-0.810)
|
77.8
|
78.7
|
0.565
|
FDP
|
5.000 mg/L
|
0.718 (0.599-0.837)
|
70.4
|
86.0
|
0.564
|
Abbreviation: VTE: Stomach cancer; t-PAIC: Tissue plasminogen activator inhibitor complex; TAT: Thrombin-antithrombin complex; TM: Thrombomodulin; PIC:α2-plamininhibitor-plasmin complex; D-D: D-Dimer; FDP: Fibrinogen degradation product
*P < 0.05 considered statistically significant.
Table 12 Receiver operating characteristic curve (ROC) analysis of 6 markers combined with 6 markers for the diagnosis of VTE
Markers
|
AUC (95%CI)
|
Sensitivity (%)
|
Specificity (%)
|
Youden Index
|
t-PAIC+TM+TAT+PIC
|
0.894 (0.815-0.972)
|
81.5
|
91.3
|
0.728
|
t-PAIC+TM+TAT+PIC+D-D+FDP
|
0.912 (0.845-0.980)
|
85.2
|
90.7
|
0.759
|
D-D+FDP
|
0.688 (0.566-0.810)
|
51.9
|
88.7
|
0.406
|