Clinicopathological Characteristics of the AUS/FLUS Nodules
Of the 138 AUS/FLUS nodules in 129 patients, 92 were pathologically confirmed as malignant, and 46 were benign. Malignant nodules included 89 PTCs, 2 FTCs, and 1 medullary thyroid carcinoma. Benign nodules included 25 nodular goiters, 5 adenomatous goiters, 4 follicular adenomas, 4 subacute thyroiditis nodules, 4 atypical hyperplasia nodules, 2 lymphocytic thyroiditis nodules, and 2 Hürthle cell adenomas (Table 2).
Table 2
Final pathologic of AUS/FLUS nodules
Postoperative pathology
|
Total
(n = 138)
|
BRAFV600E (+)
(n = 67)
|
BRAFV600E (-)
(n = 71)
|
Benign (n = 46)
|
|
|
|
Nodular goiter
|
25
|
0
|
25
|
Adenomatous goiter
|
5
|
1
|
4
|
Follicular adenoma
|
4
|
0
|
4
|
Subacute thyroiditis
|
4
|
0
|
4
|
Atypical hyperplasia
|
4
|
2
|
2
|
Lymphocytic thyroiditis
|
2
|
0
|
2
|
Hürthle cell adenoma
|
2
|
0
|
2
|
Malignant (n = 92)
|
|
|
|
PTC
|
89
|
64
|
25
|
FTC
|
2
|
0
|
2
|
MTC
|
1
|
0
|
1
|
AUS/FLUS, Atypia of Undetermined Significance or Follicular Lesion of Undetermined Significance; PTC, papillary thyroid carcinoma; FTC, follicular thyroid carcinoma; MTC, medullary thyroid carcinoma. |
The patients with malignant nodule comprised 19 men and 73 women (mean age, 45.16 ± 11.86 years; age range, 18–74 years), while the patients with benign nodule comprised 6 men and 40 women (mean age, 48.33 ± 12.64 years; age range, 19–76 years). There were no statistically significant differences in age, sex, location, and background of lymphocytic thyroiditis between the two groups (P > 0.05). There was a statistical difference in the size of benign and malignant AUS/FLUS nodules (15.9 [6.8–20.1]mm vs. 10.2 [7.0–12.0]mm, P = 0.014). Malignant nodules were more commonly detected in nodules with a size less than 10mm (P = 0.033) (Table 3).
Table 3
Clinicopathological characteristics of benign and malignant AUS/FLUS nodules
Parameter
|
Pathological findings
|
p-value
|
Benign (n = 46)
|
Malignant (n = 92)
|
Age (years), Mean ± SD
|
48.33 ± 12.64
|
45.16 ± 11.86
|
0.15
|
Sex
|
|
|
0.274
|
Male
|
6 (13.0%)
|
19 (20.7%)
|
|
Female
|
40 (87.0%)
|
73 (79.3%)
|
|
Nodule size on US (mm)
|
|
|
|
Median (P25–P75)
|
15.9 (6.8–20.1)
|
10.2 (7.0–12.0)
|
0.014
|
Size group
|
|
|
0.033
|
≤ 10
|
15 (32.6%)
|
44 (47.8%)
|
|
10–20
|
20 (43.5%)
|
40 (43.5%)
|
|
> 20
|
11 (23.9%)
|
8 (8.7%)
|
|
Location
|
|
|
0.346
|
Upper region
|
13 (28.3%)
|
31 (33.7%)
|
|
Middle region
|
14 (30.4%)
|
32 (34.8%)
|
|
Lower region
|
17 (37.0%)
|
21 (22.8%)
|
|
Isthmus region
|
2 (4.3%)
|
8 (8.7%)
|
|
Lymphocytic thyroiditis
|
|
|
0.695
|
Negative
|
33 (71.7%)
|
63 (68.5%)
|
|
Positive
|
13 (28.3%)
|
29 (31.5%)
|
|
AUS/FLUS, Atypia of Undetermined Significance or Follicular Lesion of Undetermined Significance; US, Ultrasound. |
Correlation Between the BRAFV600E Mutation and Pathological Findings
Among 138 AUS/FLUS nodules, BRAFV600E mutation was observed in 67 nodules, including 64 malignant nodules (64 PTCs) and 3 benign nodules (2 atypical hyperplasia nodules and 1 adenomatous goiter). Negative BRAFV600E mutation was identified in other 71 AUS/FLUS nodules, including 28 malignant nodules and 43 benign nodules (Table 2).
Correlation Between the C-TIRADS and Pathological Findings
The overall distribution of the C-TIRADS for all AUS/FLUS nodules was as follows: C-TIRADS 3 in 6, C-TIRADS 4A in 18, C-TIRADS 4B in 47, C-TIRADS 4C in 65, and C-TIRADS 5 in 2. The malignancy rates in C-TIRADS 3, 4A, 4B, 4C, and 5 were 0%, 33.3%, 57.4%, 87.7%, and 100%, respectively (P = 0.000), and the malignancy rate increased from C-TIRADS 3 nodules to C-TIRADS 5 nodules (P for trend = 0.000) (Table 4).
Table 4
Malignancy rates of AUS/FLUS nodules according to the C-TIRADS
|
C-TIRADS
|
p-value
|
|
3
|
4A
|
4B
|
4C
|
5
|
|
All
|
|
|
|
|
|
0.000
|
Benign
|
6
|
12
|
20
|
8
|
0
|
|
Malignant
|
0
|
6
|
27
|
57
|
2
|
|
Risk of Malignancy (%)
|
0
|
33.3
|
57.4
|
87.7
|
100
|
|
P for Trend
|
0.000
|
|
Negative BRAFV600E mutation
|
|
|
|
|
|
0.000
|
Benign
|
6
|
11
|
19
|
7
|
0
|
|
Malignant
|
0
|
3
|
5
|
18
|
2
|
|
Risk of Malignancy (%)
|
0
|
21.4
|
20.8
|
70.8
|
100
|
|
P for Trend
|
0.000
|
|
AUS/FLUS, Atypia of Undetermined Significance or Follicular Lesion of Undetermined Significance; C-TIRADS, Chinese Thyroid Imaging Reporting and Data System. |
For AUS/FLUS nodules without BRAFV600E mutation, C-TIRADS was classified as C-TIRADS 3 in 6, C-TIRADS 4A in 14, C-TIRADS 4B in 24, C-TIRADS 4C in 25, and C-TIRADS 5 in 2, respectively. The malignant rates of C-TIRADS 3, 4A, 4B, 4C, and 5 were 0%, 21.4%, 20.8%, 70.8%, and 100%, respectively (P = 0.000), and the malignancy rate increased from C-TIRADS 3 nodules to C-TIRADS 5 nodules (P for trend = 0.000) (Table 4).
The rate of nodule with BRAFV600E mutation in each C-TIRADS was as follows: C-TIRADS 3 in 0, C-TIRADS 4A in 4, C-TIRADS 4B in 23, C-TIRADS 4C in 40, and C-TIRADS 5 in 0, respectively, and the rate of nodule with BRAFV600E mutation increased from C-TIRADS 3 nodules to C-TIRADS 4C nodules (P for trend = 0.001).
Diagnostic Performance of the BRAFV600E Mutation, C-TIRADS, and the Combination
The sensitivity, specificity, PPV, NPV, accuracy, and AUC of BRAFV600E mutation were 69.6%, 93.5%, 95.5%, 60.6%, 77.5%, and 0.815, respectively. The best cut-off of C-TIRADS was C-TIRADS 4C according to the ROC curve. The sensitivity, specificity, PPV, NPV, accuracy, and AUC of C-TIRADS were 64.1%, 82.6%, 88.1%, 53.5%, 70.3%, and 0.783, respectively, and the diagnose performance was similar to the BRAFV600E mutation (P > 0.05). The sensitivity, specificity, PPV, NPV, accuracy, and AUC of the combination were 91.3%, 78.3%, 89.4%, 81.8%, 87.0%, and 0.870, respectively. The sensitivity, NPV, and accuracy of the combination were significantly higher than BRAFV600E gene or C-TIRADS alone (all P < 0.05) (Table 5, Fig. 2). Among them, a total of 18 AUS/FLUS nodules were misdiagnosed, including 10 false-positive nodules (1 benign nodule showed BRAFV600E mutation and C-TIRADS 4C, 2 showed BRAFV600E mutation, and 7 showed C-TIRADS 4C) and 8 false-negative nodules (3 malignant nodules showed C-TIRADS 4A, and 5 showed C-TIRADS 4B) (Table 6).
Table 5
Comparison of the diagnostic performance of BRAFV600E gene, C-TIRADS, and their combination
Different diagnostic means
|
Sensitivity (%)
(95% CI)
|
Specificity (%)
(95% CI)
|
PPV (%)
(95% CI)
|
NPV (%)
(95% CI)
|
Accuracy (%)
(95% CI)
|
AUC
(95% CI)
|
BRAFV600E gene
|
69.6
|
93.5
|
95.5
|
60.6
|
77.5
|
0.815
|
|
59.0–78.5
|
81.1–98.3
|
86.6–98.8
|
48.2–71.7
|
66.5–86.8
|
0.740–0.876
|
C-TIRADS
|
64.1
|
82.6
|
88.1
|
53.5
|
70.3
|
0.783
|
|
53.4–73.7
|
68.0–91.7
|
73.7–94.3
|
41.4–65.3
|
61.2–71.3
|
0.705–0.849
|
BRAFV600E + C-TIRADS
|
91.3
|
78.3
|
89.4
|
81.8
|
87.0
|
0.848
|
|
83.1–95.9
|
63.2–88.5
|
80.9–94.5
|
66.8–91.3
|
77.2–95.4
|
0.777–0.903
|
C-TIRADS, Chinese Thyroid Imaging Reporting and Data System; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the receiver operating characteristic curve; CI, confidence interval. |
Table 6
Characteristics of patients who showed false-positive or false-negative by BRAFV600E gene combined with C-TIRADS
NO
|
Age (years)
|
Sex
|
US size
(mm)
|
C-TIRADS
|
BRAFV600E status
|
Histology
|
1
|
51
|
F
|
21
|
4B
|
Positive
|
|
Atypical hyperplasia
|
2
|
70
|
F
|
17
|
4A
|
Positive
|
|
Adenomatous goiter
|
3
|
37
|
F
|
9
|
4C
|
Positive
|
|
Atypical hyperplasia
|
4
|
57
|
F
|
4
|
4C
|
|
Negative
|
Atypical hyperplasia
|
5
|
32
|
F
|
6
|
4C
|
|
Negative
|
Atypical hyperplasia
|
6
|
57
|
F
|
47
|
4C
|
|
Negative
|
Nodular goiter
|
7
|
73
|
F
|
19
|
4C
|
|
Negative
|
Subacute thyroiditis
|
8
|
41
|
F
|
19
|
4C
|
|
Negative
|
Adenomatous goiter
|
9
|
37
|
F
|
16
|
4C
|
|
Negative
|
Adenomatous goiter
|
10
|
31
|
F
|
5
|
4C
|
|
Negative
|
Adenomatous goiter
|
11
|
38
|
F
|
6
|
4B
|
|
Negative
|
PTC
|
12
|
41
|
F
|
7
|
4B
|
|
Negative
|
PTC
|
13
|
48
|
F
|
36
|
4B
|
|
Negative
|
PTC
|
14
|
55
|
F
|
6
|
4B
|
|
Negative
|
PTC
|
15
|
68
|
F
|
7
|
4A
|
|
Negative
|
PTC
|
16
|
26
|
F
|
11
|
4B
|
|
Negative
|
PTC
|
17
|
22
|
F
|
33
|
4A
|
|
Negative
|
FTC
|
18
|
43
|
F
|
26
|
4A
|
|
Negative
|
FTC
|
US, Ultrasound; C-TIRADS, Chinese Thyroid Imaging Reporting and Data System; PTC, papillary thyroid carcinoma; FTC, follicular thyroid carcinoma. |