NTRKs gene was commonly suppressed by DNA methylation in CRC
Using the methylation profile we previously generated through 450K microarray, we found NTRKs gene, including NTRK1, NTRK2 and NTRK3, had more frequently methylated promoters in CRC samples when compared with matched normal mucosae (NTRK1, cancer = 0.444, normal = 0.397, P = 0.012; NTRK2, cancer = 0.251, normal = 0.167, P < 0.001; NTRK3, cancer = 0.395, normal = 0.144, P < 0.001; Fig. 2; Table S2). In support of methylation analysis results, we found the mRNA expression of NTRK2 and NTRK3 in CRC samples was commonly lower than that in normal mucosae using the expression profiles in five CRC cohorts (n = 1460). However, no significant difference was observed in NTRK1 mRNA expression (Fig. 2). Moreover, the lower expression of NTRKs gene is associated with their mean methylation of promoter (NTRK1, P = 0.049, R2 = 0.011; NTRK2, P < 0.001, R2 = 0.115; NTRK3, P < 0.001, R2 = 0.091; Fig. 2).
A specific methylated region within NTRK3 promoter best predicted CRC death
We next sought to identify the CpG site that could best predict CRC death and be feasibly used in a clinical assay. In the Cox proportional hazards analyses of 450K microarray probes targeting genomic loci within NTRKs gene, we found that the methylation of most CpG sites targeted by these probes was associated with poor survival outcomes in CRC (Fig. 2D, Table S3). Among them, the methylation of cg27034819 was top-ranked for predicting CRC death. Of note, we found the probe cg11525479 was very close to cg27034819 in their targeting loci (Fig. 2E), and the methylation of cg11525479 also had a predictive value for CRC death that was superior to most probes. These results suggested that the specific region within NTRK3 promoter targeted by cg27034819 and cg11525479 could be used to stratify the death risk of CRC. Interestingly, the methylation of this specific region was shown to be associated with the loss of NT3-dependent tumor suppressor gene function of NTRK3 in our previous in vitro and in vivo study (10). Therefore, a QMSP assay for determining the methylation of this specific region within NTRK3 promoter was developed (Fig. 2E).
Cohort validation of NTRK3 methylation for prognostic significance
We further validated the predictive value of this candidate region within NTRK3 promoter in our institutional cohort. The baseline characteristics of this validation cohort were summarized in Table 1. NTRK3 promoter hypermethylation was observed in 26 of 229 patients (11.35%), and it was more frequent in patients with MSI (P = 0.015; Table 1). Moreover, NTRK3 promoter hypermethylation was associated with KRAS mutation (P = 0.001; Table 1). For other characteristics relevant to clinical outcomes of CRC, they did not show significant difference between patients with NTRK3 promoter hypermethylation and hypomethylation, including age, sex, tumor size, tumor differentiation, lymphovascular invasion, perineural invasion, TNM stage, CIMP, MSI, KRAS mutation, BRAF mutation, Ki-67, CA19-9, and CEA.
Table 1
Baseline characteristics and NTRK3 promotor methylation status among all patients with CRC in validation cohort
|
Total
|
NTRK3 promotor methylation
|
P value
|
Variables
|
hypomethylation
|
hypermethylation
|
Age (median = 62 years)
|
|
|
|
|
< 62
|
111
|
100
|
11
|
|
≥ 62
|
118
|
103
|
15
|
0.504
|
Sex
|
|
|
|
|
Female
|
100
|
88
|
12
|
|
Male
|
129
|
115
|
14
|
0.786
|
Tumor localization
|
|
|
|
|
Colon
|
126
|
114
|
12
|
|
Rectum
|
103
|
89
|
14
|
0.334
|
Tumor size (median = 4.5 cm)
|
|
|
|
|
< 4.5
|
104
|
90
|
14
|
|
≥ 4.5
|
123
|
111
|
12
|
0.382
|
unknown
|
2
|
|
|
|
Tumor differentiation
|
|
|
|
|
High or moderate
|
191
|
168
|
23
|
|
Poor
|
38
|
35
|
3
|
0.584
|
Lymphovascular invasion
|
|
|
|
|
No
|
210
|
188
|
22
|
|
Yes
|
17
|
13
|
4
|
0.219
|
unknown
|
2
|
|
|
|
Perineural invasion
|
|
|
|
|
No
|
208
|
183
|
25
|
|
Yes
|
19
|
18
|
1
|
0.611
|
unknown
|
2
|
|
|
|
TNM stage
|
|
|
|
|
I-II
|
143
|
123
|
20
|
|
III-IV
|
85
|
79
|
6
|
0.112
|
unknown
|
1
|
|
|
|
CIMP
|
|
|
|
|
negative
|
222
|
198
|
24
|
|
Positive
|
7
|
5
|
2
|
0.145
|
Microsatellite status
|
|
|
|
|
MSS
|
125
|
113
|
12
|
|
MSI
|
51
|
39
|
12
|
0.015
|
unknown
|
53
|
|
|
|
KRAS
|
|
|
|
|
wild-type
|
119
|
110
|
9
|
|
mutation
|
62
|
46
|
16
|
0.001
|
unknown
|
48
|
|
|
|
BRAF
|
|
|
|
|
wild-type
|
174
|
151
|
23
|
|
mutation
|
8
|
6
|
2
|
0.674
|
unknown
|
47
|
|
|
|
Ki-67
|
|
|
|
|
≤ 25%
|
100
|
89
|
11
|
|
> 25%
|
84
|
72
|
12
|
0.502
|
unknown
|
45
|
|
|
|
CA19-9
|
|
|
|
|
≤ 37
|
179
|
160
|
19
|
|
> 37
|
32
|
29
|
3
|
0.833
|
unknown
|
18
|
|
|
|
CEA
|
|
|
|
|
≤ 5
|
158
|
138
|
20
|
|
> 5
|
57
|
54
|
3
|
0.122
|
unknown
|
14
|
|
|
|
CIMP, CpG island methylator phenotype; KRAS, kirsten rat sarcoma viral oncogene; BRAF, B-Raf proto-oncogene, serine/threonine kinase; Ki-67, kiel67 antigen; CA19-9, carbohydrate antigen 19 − 9; CEA, carcinoembryonic antigen. |
In the Kaplan-Meier curve, significantly worse DFS outcomes were observed in patients with NTRK3 promoter hypermethylation compared to those with NTRK3 promoter hypomethylation (P = 0.012; Fig. 3A). The prognostic value of NTRK3 promoter methylation status was further confirmed by univariate Cox proportional hazards (P = 0.014, HR = 2.194, 95% CI = [1.169, 4.117]; Table 2). Next, in the light of multivariate analysis, NTRK3 promoter hypermethylation was still a prognostic factor adjusted by age, TNM stage, and BRAF mutation (P = 0.004, HR = 2.688, 95% CI = [1.355, 5.333]; Table 2).
Table 2
Cox Proportional Hazard analyses on DFS in patients with CRC
Variables
|
DFS in the patients with I-IV stage CRC
|
DFS in the patients with I-III stage CRC
|
Univariate
|
Multivariate
|
Univariate
|
Multivariate
|
P value
|
HR[95%CI]
|
P value
|
HR[95%CI]
|
P value
|
HR[95%CI]
|
P value
|
HR[95%CI]
|
NTRK3 hypermethylation
|
0.014
|
2.194(1.169, 4.117)
|
0.004
|
2.688(1.355, 5.333)
|
0.004
|
2.565(1.354, 4.857)
|
0.015
|
2.630(1.206, 5.734)
|
Older age
|
0.011
|
1.952(1.168, 3.263)
|
0.060
|
1.723(0.977, 3.038)
|
0.002
|
2.401(1.364, 4.227)
|
0.010
|
2.334(1.217, 4.477)
|
Male (vs. female)
|
0.574
|
1.153(0.702, 1.894)
|
|
|
0.484
|
1.207(0.713, 2.043)
|
|
|
Rectal tumor (vs. colon tumor)
|
0.479
|
1.192(0.733, 1.939)
|
|
|
0.341
|
1.087(0.915, 1.291)
|
|
|
Tumor size ≥ 4.5
|
0.056
|
1.655(0.988, 2.772)
|
|
|
0.081
|
1.624(0.943, 2.799)
|
|
|
Poor differentiation
|
0.581
|
1.193(0.637, 2.233)
|
|
|
0.553
|
1.220(0.633, 2.353)
|
|
|
Lymphovascular invasion
|
0.052
|
2.087(0.995, 4.378)
|
|
|
0.022
|
2.387(1.131, 5.038)
|
0.003
|
3.287(1.465, 7.373)
|
Perineural invasion
|
0.350
|
1.455(0.663, 3.195)
|
|
|
0.158
|
1.770(0.801, 3.912)
|
|
|
Advanced TNM stage
|
0.003
|
2.125(1.300, 3.473)
|
< 0.001
|
2.704(1.528, 4.790)
|
0.019
|
1.854(1.107, 3.103)
|
0.039
|
1.978(1.034, 3.784)
|
CIMP positive status
|
0.265
|
1.935(0.607, 6.169)
|
|
|
0.186
|
2.191(0.685, 7.010)
|
|
|
MSI
|
0.316
|
1.350(0.751, 2.427)
|
|
|
0.138
|
1.599(0.860, 2.973)
|
|
|
KRAS mutation
|
0.113
|
1.566(0.899, 2.726)
|
|
|
0.067
|
1.743(0.962, 3.157)
|
|
|
BRAF mutation
|
0.010
|
3.376(1.338, 8.519)
|
0.049
|
2.563(1.004, 6.551)
|
0.003
|
4.122(1.621, 10.485)
|
0.595
|
1.443(0.373, 5.591)
|
Ki-67 > 25%
|
0.458
|
0.806(0.456, 1.425)
|
|
|
0.675
|
0.877(0.476, 1.617)
|
|
|
CA19-9 > 37
|
0.078
|
1.780(0.938, 3.377)
|
|
|
0.191
|
1.589(0.794, 3.179)
|
|
|
CEA > 5
|
0.095
|
1.591(0.923, 2.741)
|
|
|
0.048
|
1.769(1.005, 3.111)
|
0.772
|
1.122(0.514, 2.447)
|
NTRK3, neurotropic tropomyosin receptor kinase 3; CIMP, CpG island methylator phenotype; MSI, microsatellite instability; KRAS, kirsten rat sarcoma viral oncogene; BRAF, B-Raf proto-oncogene, serine/threonine kinase; Ki-67, kiel67 antigen; CA19-9, carbohydrate antigen 19 − 9; CEA, carcinoembryonic antigen. |
In sensitivity analyses, NTRK3 promoter methylation was still independently associated with poor DFS outcome after the exclusion of patients with stage IV disease (P = 0.015, HR = 2.630, 95% CI = [1.206, 5.734]; Fig. 3B, Table 2), CIMP (P = 0.003, HR = 2.806, 95% CI = [1.432, 5.500]; Table S4), MSI (P = 0.008, HR = 3.483, 95% CI = [1.391, 8.717]; Table S5), or BRAF mutation (P = 0.025, HR = 2.603, 95% CI = [1.125, 6.022]; Table S6) in multivariate Cox analysis.
A Nomogram For Predicting DFS In CRC Patients
A nomogram for predicting 3-year and 5-year DFS outcome was generated using the variables from the multivariate Cox model, including NTRK3 methylation, age at diagnosis, TNM stage, and BRAF mutation (Fig. 3C). The calibration curves for the nomogram were shown(Fig. 3D). The C-index of the nomogram for predicting DFS was 0.705.
NTRK3 methylation adds values to current prognostic panels
The model 1 had a lower AIC and a higher LR compared with the model 2(AIC: 597.73 vs 600.69; LR: 6.91 vs 5.95, P = 0.005; Table 3), indicating that NTRK3 hypermethylation alone is better in predicting prognosis than rough TNM staging alone. In the comparison between model 2 and 3, after NTRK3 hypermethylation was added to TNM stage, a lower AIC and a higher LR were observed (AIC: 600.69 vs 592.41; LR: 5.95 vs 16.23, P = 0.002; Table 3). These results suggest NTRK3 hypermethylation could increase prognostic values of TNM staging.
Table 3
Model fit among seven models including or not including NTRK3 methylation status
Models
|
N
|
AIC
|
LR
|
P value
|
Model 1
|
219
|
597.73
|
6.91
|
|
Model 2
|
219
|
600.69
|
5.95
|
0.005a
|
Model 3
|
219
|
592.41
|
16.23
|
0.002b
|
Model 4
|
219
|
516.49
|
39.06
|
|
Model 5
|
219
|
513.91
|
43.64
|
0.032c
|
N, patient counts in each model; AIC, Akaike information criterion value; LR, likelihood ratio. |
Model 1 includes NTRK3 hypermethylation. |
Model 2 includes TNM stage (I, II, III). |
Model 3 includes TNM stage (I, II, III), NTRK3 hypermethylation. |
Model 4 includes age at diagnosis, sex (m/f), location(left colon, right colon, rectum), differentation(well, moderate, poor), lymphovascular invasion, perineural invasion(n/y), T stage(T1, T2, T3, T4), N stage(N0, N1, N2), negative lymph node number, preoperation CEA, chemotherapy(n/y). |
Model 5 includes NTRK3 hypermethylation and all variables in Model 4. |
a, P values for the LR test in model 1 compared with model 2; b, P values for the LR test in model 2 compared with model 3; c, P values for the LR test in model 4 compared with model 5. |
To determine the values of NTRK3 hypermethylation in commonly-used models using multiple clinicopathological variables, model 4 was built using the variables included in the model recommended by AJCC(43, 44). As expected, after NTRK3 hypermethylation was included, model 5 had a lower AIC and a higher LR in comparison to model 4 (AIC: 516.49 vs 513.91; LR: 39.06 vs 43.64, P = 0.032; Table 3). Thus, the model recommended by AJCC may get increased discriminatory ability in predicting prognosis with NTRK3 hypermethylation.
Prognostic significance of NTRK3 methylation in multiple tumors
Both the methylation of cg27034819 and cg11525479 were analyzed on their associations with survival outcome in 23 tumors using TCGA methylation profiles generated by 450K microarray. Overall, similar to the conflicting findings from in vitro and in vivo studies on NTRKs gene, the association of their methylations with survival outcome varied in different tumors. The methylation of cg27034819 was significantly associated with worse survival outcome in COAD (P = 0.008, HR = 1.91, 95% CI = [1.18, 3.09]), READ (P = 0.006, HR = 3.73, 95% CI = [1.37, 10.15]), KICH (P = 0.005, HR = 7.36, 95% CI = [1.83, 29.581], and PAAD (P = 0.001, HR = 1.96, 95% CI = [1.30, 2.96]) cohorts. However, it was significantly associated with better survival outcome in GBM (P = 0.037, HR = 0.63, 95% CI = [0.41, 0.97], SKCM (P = 0.002, HR = 0.58, 95% CI = [0.41, 0.81]) and STAD (P = 0.013, HR = 0.57, 95% CI = [0.37, 0.89]) cohorts (Fig. 3E). In cg11525479 methylation analysis, a similar predictive value for worse survival were found in COAD, READ, and PAAD, and a similar predictive value for better survival were found in GBM, SKCM and STAD (Fig. 3F). These results suggested a robust prognostic value of the methylation of the specific promoter region targeted by cg27034819 and cg11525479 in multiple tumors.