The Relationship Between CTCs Expression Levels and Survival in Children with NB
By the follow-up deadline in June 2024, 43 children were still alive, with a survival rate of 29.86%. The average survival time of the children was 21.72 months (95% CI: 20.23~23.20). The prognostic value of CTCs in NB was analyzed using X-tile software. The results showed that when the optimal cut-off value was <14.48, the prognosis of the children significantly improved (Figure 1A-B).
Comparison of Baseline Data Based on CTCs Expression Levels
According to the optimal cut-off values obtained from X-tile software, the children were divided into high and low expression groups. Further comparison of the baseline data between the two groups revealed that the number of children with Progress (P<0.001), Maximum tumor diameter ≥10cm (P=0.004), low differentiation (P=0.034), and stage IV (P=0.007) in the high expression group was significantly higher than in the low expression group, indicating statistical significance (Table 1). There was no statistical significance in age, sex, and risk level between the groups (P>0.05).
Table 1 Relationship Between CTCs Expression Levels and Baseline Data
Factors
|
|
CTCs Expression Levels
|
χ2 Value
|
P Value
|
High Expression (n=69)
|
Low Expression (n=75)
|
Age
|
|
|
|
|
|
|
≥18 months
|
31
|
43
|
2.214
|
0.137
|
|
<18 months
|
38
|
32
|
|
|
Sex
|
|
|
|
|
|
|
Male
|
43
|
53
|
1.127
|
0.288
|
|
Female
|
26
|
22
|
|
|
Efficacy
|
|
|
|
|
|
|
Progress
|
40
|
17
|
18.73
|
<0.001
|
|
Mitigation
|
29
|
58
|
|
|
Maximum tumor diameter
|
|
|
|
|
|
|
≥10cm
|
51
|
38
|
8.227
|
0.004
|
|
<10cm
|
18
|
37
|
|
|
Degree of differentiation
|
|
|
|
|
|
|
Low differentiation
|
29
|
19
|
4.508
|
0.034
|
|
Medium + high differentiation
|
40
|
56
|
|
|
Tumor stage
|
|
|
|
|
|
|
Stage IV
|
43
|
30
|
7.162
|
0.007
|
|
Stage III
|
26
|
45
|
|
|
Risk level
|
|
|
|
|
|
|
High risk
|
52
|
53
|
0.401
|
0.526
|
|
Medium risk
|
17
|
22
|
|
|
Note: Circulating Tumor Cells (CTCs).
CTCs Expression in Efficacy, Maximum Tumor Diameter, Degree of Differentiation, and Tumor Stage
Baseline data analysis revealed that CTCs were associated with Efficacy, Maximum tumor diameter, Degree of differentiation, and Tumor stage. Therefore, further comparison of CTCs expression levels in different baseline data was conducted. The results showed that CTCs in children with Progress were higher than those in children with Mitigation, with statistical significance (P<0.001, Figure 2A); children with Maximum tumor diameter <10cm had lower CTCs than those with ≥10cm, with statistical significance (P<0.001, Figure 2B); there was no difference in CTCs among children with different degrees of differentiation (P=0.148, Figure 2C); children with stage IV had higher CTCs than those with stage III, with statistical significance (P=0.036, Figure 2D).
Diagnostic Value of CTCs in Efficacy, Maximum Tumor Diameter, Degree of Differentiation, and Tumor Stage
The above study determined that CTCs expression was related to Efficacy, Maximum tumor diameter, Degree of differentiation, and Tumor stage. Therefore, separate ROC curves for CTCs in Efficacy, Maximum tumor diameter, Degree of differentiation, and Tumor stage were plotted. The results showed that the area under the curve (AUC) for Efficacy, Maximum tumor diameter, Degree of differentiation, and Tumor stage were 0.703, 0.669, 0.574, and 0.598, respectively (Figure 3, Table 2). Among them, CTCs had the highest AUC in distinguishing clinical efficacy, while the AUC for distinguishing tumor stage and degree of differentiation was the lowest.
Table 2 ROC Parameters
Metric
|
Efficacy
|
Maximum tumor diameter
|
Degree of differentiation
|
Tumor stage
|
AUC
|
0.703
|
0.669
|
0.574
|
0.598
|
95%CI
|
0.614 - 0.703
|
0.577 - 0.669
|
0.476 - 0.574
|
0.504 - 0.598
|
specificity
|
0.724
|
0.491
|
0.385
|
0.718
|
sensitivity
|
0.684
|
0.809
|
0.833
|
0.521
|
Youden index
|
0.408
|
0.3
|
0.219
|
0.239
|
Cut off
|
14.845
|
11.655
|
11.755
|
15.06
|
Accuracy
|
0.708
|
0.688
|
0.535
|
0.618
|
Precision
|
0.619
|
0.72
|
0.404
|
0.655
|
F1 Score
|
0.65
|
0.762
|
0.544
|
0.58
|
Note: Circulating Tumor Cells (CTCs) and Area Under Curve (AUC).
Prognostic Factors Affecting OS in Children with NB
In this study, Cox regression analysis was used to screen prognostic factors affecting OS in children with NB. Univariate analysis showed that CTCs (P<0.001, OR=4.221), Efficacy (P<0.001, OR=2.395), Maximum tumor diameter (P<0.001, OR=3.43), and Degree of differentiation (P=0.01, OR=1.69) were prognostic factors affecting the prognosis of children (Table 3). Multivariate Cox regression analysis showed that CTCs (P<0.001, OR=2.958), Efficacy (P=0.005, OR=1.814), and Maximum tumor diameter (P<0.001, OR=2.586) were independent prognostic factors affecting OS in children with NB (Table 4, Figure 4).
Table 3 Univariate Cox Regression
Factor
|
Beta
|
StdErr
|
PValue
|
HR
|
95% CI Lower Limit
|
95% CI Upper Limit
|
CTCs
|
1.440
|
0.213
|
<0.001
|
4.221
|
2.782
|
6.404
|
Age
|
0.060
|
0.199
|
0.763
|
1.062
|
0.718
|
1.570
|
Sex
|
-0.041
|
0.211
|
0.847
|
0.960
|
0.635
|
1.451
|
Efficacy
|
0.873
|
0.200
|
<0.001
|
2.395
|
1.617
|
3.548
|
Maximum tumor diameter
|
1.232
|
0.237
|
<0.001
|
3.430
|
2.154
|
5.461
|
Degree of differentiation
|
0.525
|
0.205
|
0.010
|
1.690
|
1.131
|
2.525
|
Tumor stage
|
0.296
|
0.200
|
0.138
|
1.345
|
0.909
|
1.990
|
Risk level
|
-0.008
|
0.222
|
0.973
|
0.992
|
0.642
|
1.534
|
Note: Circulating Tumor Cells (CTCs).
Table 4 Multivariate Cox Regression
Factor
|
Beta
|
StdErr
|
PValue
|
HR
|
95% CI Lower Limit
|
95% CI Upper Limit
|
CTCs
|
1.084
|
0.230
|
<0.001
|
2.958
|
1.884
|
4.643
|
Efficacy
|
0.595
|
0.211
|
0.005
|
1.814
|
1.199
|
2.744
|
Maximum tumor diameter
|
0.950
|
0.250
|
<0.001
|
2.586
|
1.584
|
4.222
|
Degree of differentiation
|
0.166
|
0.210
|
0.428
|
1.181
|
0.783
|
1.781
|
Note: Circulating Tumor Cells (CTCs).
Construction of a Nomogram Model Based on Independent Prognostic Factors
At the end of the study, a visual Nomogram model was constructed based on independent prognostic factors. The model included Efficacy, Maximum tumor diameter, and CTCs. Among them, CTCs were strongly correlated with patients' OS, while Efficacy and Maximum tumor diameter were related to tumor OS (Figure 5A). Furthermore, time-dependent ROC curve analysis showed that the model's AUC for predicting 1-year and 2-year survival was 0.747 and 0.802, respectively, indicating that the model has certain clinical value in predicting patients' survival (Figure 5B).