Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.
Original Research n Ultrasonography
Association of Preoperative US Features and Recurrence in Patients with Classic Papillary Thyroid Carcinoma1 Soo-Yeon Kim, MD Jin Young Kwak, MD, PhD Eun-Kyung Kim, MD, PhD Jung Hyun Yoon, MD Hee Jung Moon, MD, PhD
Purpose:
To investigate whether ultrasonographic (US) features of thyroid nodules are associated with tumor recurrence in classic papillary thyroid carcinoma (PTC).
Materials and Methods:
This retrospective study was approved by the Institutional Review Board, and the need to obtain informed consent was waived. A total of 515 patients (mean age, 45.8 years 6 13.2 [standard deviation]; range, 17–80 years) who underwent total thyroidectomy and central lymph node dissection for classic PTC greater than 10 mm from January 2003 to February 2006 and who were followed up for 12 months or longer were included. Malignant-appearing PTCs were defined as those showing at least one suspicious US feature among marked hypoechogenicity, irregular or microlobulated margin, microcalcifications, and taller-than-wide shape. Benign-appearing PTCs were defined as those without any suspicious US features. Kaplan-Meier cumulative-event curves for recurrence were compared by using the log-rank test. The multivariate Cox proportional hazard regression analysis was used to estimate hazard ratios (HRs) of the malignant-appearing US features for recurrence in the preoperative, postoperative, and combined models.
Results:
Fifty-six (10.9%) of 515 patients had recurrence. Malignant-appearing PTCs had lower 5- and 10-year diseasefree survival rates compared with benign-appearing PTCs (P = .01). In the preoperative model, malignant-appearing US features (HR, 3.523; 95% confidence interval [CI]: 1.263, 9.830) and larger nodule size (HR, 1.074; 95% CI: 1.051, 1.098) were independently associated with recurrence. In the combined model, male sex (HR, 1.990; 95% CI: 1.098, 3.610), malignant-appearing US features (HR, 2.828; 95% CI: 1.016. 7.870), larger nodule size (HR, 1.067; 95% CI: 1.043, 1.092), extrathyroidal extension (HR, 2.590; 95% CI: 1.160, 5.780), and lymph node metastasis (HR, 2.511; 95% CI: 1.163, 5.421) were independently associated with recurrence.
Conclusion:
The presence of malignant-appearing US features was independently associated with recurrence in patients with classic PTC.
1
From the Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Yonsei University, College of Medicine, 50 Yonsei-ro, Seodaemungu, 120-752 Seoul, Korea. Received October 22, 2014; revision requested December 9; revision received January 20, 2015; accepted February 6; final version accepted February 19 Address correspondence to H.J.M. (e-mail: [email protected]).
RSNA, 2015
q
RSNA, 2015
q
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
1
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
P
apillary thyroid carcinoma (PTC) generally displays an indolent course and has favorable prognosis, with overall 10-year survival rates reported to be 93%–95% (1–4). The reported 10-year recurrence rates are about 14%–26% (2,4–9), and the independent risk factors for recurrence are age at diagnosis, tumor size, pathologic subtype, radioactive iodine therapy, extrathyroidal extension, and lymph node metastasis (2,4–18). Recently, in a report of 488 patients with PTC, ultrasonographic (US) features of PTC at the time of diagnosis served as useful prognostic factors, with poor prognostic factors being more associated with malignant-appearing PTC than benign-appearing PTC (19). Malignant-appearing PTC more frequently had extrathyroidal extension, lymph node metastasis, and a higher TNM stage than benign-appearing PTC (19). However, recurrence was not significantly different between the two groups (19). The mean follow-up time was 60 months (range, 18–71 months); thus, the authors thought that recurrence might be underreported because of the relatively short follow-up periods being observed in a certain portion of patients (19). If the presence of malignant US features is an independent prognostic factor for recurrence, preoperative US features could be used to predict prognosis in individual patients with PTC. These US features could be used to decide the extent of initial surgery, postoperative adjuvant therapy, and the intensity of postoperative surveillance.
Advance in Knowledge nn Malignant-appearing US features of classic papillary thyroid carcinoma (PTC) were significantly associated with recurrence in the preoperative model (hazard ratio, 3.523; 95% confidence interval; 1.263, 9.830; P = .016), as well as the combined model of preoperative and postoperative factors (hazard ratio, 2.828; 95% confidence interval; 1.016, 7.870; P = .047). 2
Classic PTCs 10 mm or smaller have excellent prognosis (19–23). According to a meta-analysis by Roti et al (23), the combination of results from different studies revealed a recurrence rate of 2.4%. PTCs 10 mm or smaller show a higher tendency for false-positive features at US or inadequate cytologic findings at US-guided fine-needle aspiration compared with those larger than 10 mm (24,25). Therefore, we investigated whether US features of thyroid nodules are associated with tumor recurrence in classic PTC larger than 10 mm.
Materials and Methods Patients The Institutional Review Board approved this retrospective study, and informed consent was waived. From January 2003 to February 2006, 2111 consecutive patients underwent thyroid surgery. A total of 1596 patients were excluded: (a) 1133 patients with PTC 10 mm or less, (b) 174 patients with subtotal thyroidectomy or hemithyroidectomy, (c) 150 patients with no available preoperative US images, (d) 49 patients with PTC other than the classic subtype or no PTC, (e) 46 patients with completion thyroidectomy, (f) 30 patients with malignancy other than thyroid cancer, and (g) 14 patients with a follow-up period of less than 12 months after surgery (Fig 1). We included 515 patients who underwent total thyroidectomy for classic PTC larger than 10 mm and had a follow-up period at least 12 months after surgery. Prophylactic central lymph node dissection was routinely performed at the time of total thyroidectomy for all included patients. The mean age of the 515 patients (432 women [mean age, 45.5 years 6 5.5 [standard deviation]; range, 17–80 years] and 83 men [mean age, 47.5 years 6 12.0; range, 21–74 years]) Implication for Patient Care nn The presence of malignantappearing US features of classic PTC is strongly associated with recurrence.
Kim et al
was 45.8 years 6 13.2 (range, 17–80 years). Among the 515 patients, 191 and 94 were previously reported in two published articles, respectively (26,27). However, the association between preoperative US features and recurrence of classic PTC was the primary concern of the present study, and none of these concepts were documented in the previous studies (26,27).
Image Analysis From January 2003 to February 2006, preoperative thyroid US was performed by one of six radiologists, with 1–11 years of experience in thyroid imaging, using 5–15-MHz linear array transducers (HDI 5000, Philips Medical Systems, Bothell, Wash; iU22, Philips Medical Systems; Acuson Sequoia 512 or 513, Siemens Medical Solutions, Mountain View, Calif). When we perform thyroid US in daily practice, we differentiate suspicious nodules from probably benign nodules on the basis of published criteria, such as marked hypoechogenicity, microlobulated or irregular margins, microcalcifications, and taller-than-wide shape (28,29). We report the size and location of a thyroid nodule—whether a nodule is located on (a) the right or left thyroid lobe and on (b) the upper, middle, or lower part of a thyroid lobe in the
Published online before print 10.1148/radiol.2015142470 Content codes: Radiology 2015; 000:1–10 Abbreviations: CI = confidence interval HR = hazard ratio PTC = papillary thyroid carcinoma Tg = thyroglobulin Author contributions: Guarantors of integrity of entire study, J.Y.K., H.J.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, S.Y.K., H.J.M.; clinical studies, J.Y.K., E.K.K., H.J.M.; experimental studies, H.J.M.; statistical analysis, S.Y.K., H.J.M.; and manuscript editing, E.K.K., J.H.Y., H.J.M. Conflicts of interest are listed at the end of this article.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
lymph nodes at level 2, 3, 4, and anterior 5. Therapeutic lateral compartment dissection was performed in 113 patients in our study. Tumor size, multifocality of the tumor, extrathyroidal extension, and central and/or lateral lymph node metastasis status were reassessed and recorded from original pathologic reports without researchers having knowledge of the US features. Index tumors were re-evaluated for TNM staging according to the seventh edition of the American Joint Committee on Cancer staging system and the International Union against Cancer pathologic TNM classification criteria on the basis of their pathologic reports (30).
Figure 1
Figure 1: Flowchart of the study population.
original radiologic report. Also, when we perform fine-needle aspiration, we obtain an image of the needle located within the index nodule. We match the index tumor of US with that of pathologic examination based on location and size. For this study, all US images of index tumors, which were routinely evaluated in both transverse and longitudinal planes and were scanned in both planes, were retrospectively reviewed by one of two radiologists (J.Y.K. for studies performed between January 2003 and August 2004 and H.J.M. for studies performed between September 2004 and February 2006, with 15 and 12 years of experience in thyroid imaging, respectively). There were no special intentions to segregate reviewer cohorts. The US images obtained between January 2003 to August 2004 were previously reviewed by J.Y.K. for the other studies (26,27), and those obtained between September 2004 and February 2006 were reviewed by H.J.M. for this study. The US images were reviewed according to the following five criteria: (a) internal composition (completely solid, or cystic portion , 50% and cystic portion 50% in mixed cystic and solid nodules), (b) echogenicity (hyper-, iso-, or hypoechogenicity, compared with the echogenicity of the underlying thyroid parenchyma or marked hypoechogenicity compared with adjacent strap muscle), (c) margin (well defined, microlobulated, or
Kim et al
irregular), (d) calcifications (no calcification, macrocalcification, or microcalcification [includes nodules with both micro- and macrocalcifications]), and (e) shape (wider than tall, greater in the transverse dimension than the anteroposterior dimension or taller than wide, greater in the anteroposterior dimension than the transverse dimension). The radiologists were blinded to information about the presence of recurrence in patients at the time of US review. Malignant-appearing PTCs were defined as those showing at least one suspicious US feature among marked hypoechogenicity, irregular or microlobulated margins, microcalcifications, and taller-than-wide shape, based on published criteria (28,29). Benignappearing PTCs were defined as those without any suspicious US features.
Surgery and Pathologic Diagnosis In our institution, total thyroidectomy is performed in patients aged 45 years or older with extrathyroidal extension, multifocality, or lymph node metastasis at preoperative or intraoperative examination. Prophylactic central compartment lymph node dissection is routinely performed in patients with total thyroidectomy. Lateral compartment lymph node dissection is performed only when lymph node metastasis is diagnosed at preoperative US-guided fine-needle aspiration or on an intraoperative frozen section. Lateral compartment lymph node dissection includes
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Postoperative Follow-up and Recurrence All patients underwent thyroid-stimulating hormone–suppressive therapy with levothyroxine after surgery. Radioactive iodine therapy was performed in patients with extrathyroidal extension or lymph node metastasis at histopathologic analysis. Four hundred thirtyseven (84.9%) patients underwent radioactive iodine therapy after surgery with 30 mCi (1110 Bq) administered in 420 patients and 50–200 mCi (1850– 7400 Bq) administered in 17 patients. Patients who underwent surgery were followed up every 6 months for the first 3 years and at 12-month intervals thereafter. The routine follow-up protocol consisted of a clinical examination every 6 months; the measurement of serum thyroid-stimulating hormone, free thyroxine, thyroglobulin (Tg), and anti-Tg antibody; a chest x-ray; and a neck US performed every 12 months. Other imaging examinations, such as iodine 131 whole-body scintigraphy, chest computed tomography (CT), or fluorodeoxyglucose positron emission tomography (PET)/CT, were performed in selected cases (eg, detectable or persistent serum Tg or anti-Tg antibody without evidence of recurrence on chest x-rays or neck US scans). No clinical evidence of recurrence was defined as no biochemical (suppressed Tg , 1 ng/mL, stimulated Tg , 2 ng/mL, no detectable anti-Tg antibody), no structural 3
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
(no cytology-proven disease and no morphologic evidence of disease at cross-sectional imaging), and no functional evidence of disease (no suspicious findings at whole-body scintigraphy or PET/CT) (31). Recurrence was defined as new biochemical, structural, or functional evidence of disease that was detected following any period of no evidence of disease (31). Persistent disease was defined as biochemical, structural, or functional evidence of disease that was persistently present without a period of any evidence of disease since the initial surgery.
Data and Statistical Analysis The group with recurrence included patients with recurrence according to the definitions described earlier, and the group without recurrence included patients without recurrence as well as those with persistent disease.
Disease-free survival was defined as the time interval (in months) between surgery and recurrence for the group with recurrence and the time interval between surgery and the latest clinical follow-up for the group without recurrence. We used either the x2 test or the Fisher exact test for categorical variables and the independent t test for continuous variables to compare the differences in demographic, US, and pathologic characteristics between patients with and patients without malignant-appearing US features and between patients with and patients without recurrence. Kaplan-Meier cumulative-event curves were constructed for recurrence in the two groups with benign- or malignantappearing PTC, and the 5-year and 10-year disease-free survival curves were compared with 95% confidence
Kim et al
intervals (CIs) by using the log-rank test. The univariate and multivariate Cox proportional hazard regression models were used to estimate hazard ratios (HRs) with 95% CIs for recurrence according to patient characteristics. Considering the small numbers of recurrence events, all independent variables were classified into binary categories. US features were divided into nonsuspicious and suspicious findings as follows: echogenicity (hyper-, iso-, or hypo- vs marked hypoechogenicity), margin (well defined vs microlobulated or irregular), calcifications (none or macro- vs microcalcification), shape (parallel vs nonparallel), and final assessment (benign vs malignant appearing). Internal composition was divided into completely solid or mixed cystic and solid. Pathologic characteristics were categorized according to the presence
Table 1 Characteristics of All Patients and Patients with Benign- or Malignant-appearing PTC Characteristic No. of patients Mean follow-up (mo)† Mean age (y)‡ Male sex Internal composition at US Solid ,50% cystic 50% cystic Size (mm)† Multifocality Extrathyroidal extension Central and/or lateral lymph node metastasis Central lymph node metastasis Lateral lymph node metastasis TNM stage I II III IV Radioiodine ablation therapy
All Patients
Benign- appearing PTC
Malignant- appearing PTC
515 93 6 29 45.80 6 13.15 (17–80) 83 (16.1)
113 (21.9) 87 6 26 43.58 6 13.46 (17–78) 14 (12.4)
402 (78.1) 94 6 29 46.42 6 13.01 (17–80) 69 (17.2)
473 (91.8) 36 (7.0) 6 (1.2) 19.76 6 8.76 151 (29.3) 363 (70.5) 321 (62.3)
99 (87.6) 13 (11.5) 1 (0.9) 20.92 6 9.07 24 (21.2) 69 (61.1) 59 (52.2)
374 (93.0) 23 (5.7) 5 (1.2) 19.43 6 8.65 127 (31.6) 294 (73.1) 262 (65.2)
307 (59.6) 113 (21.9)
57 (50.4) 18 (15.9)
250 (62.2) 95 (23.6)
282 (54.8) 5 (1.0) 165 (32.0) 63 (12.2) 437 (84.9)
75 (66.4) 3 (2.7) 25 (22.1) 10 (8.8) 95 (84.1)
207 (51.5) 2 (0.5) 140 (34.8) 53 (13.2) 342 (85.1)
P Value* .072 .043 .282 .100
.109 .043 .018 .016 .030 .105 .004§/.002||
.793
Note.—Unless otherwise indicated, data are the number of patients and data in parentheses are percentages. * Comparison of patients with and those without malignant-appearing PTCs.
4
†
Data are means 6 standard deviation.
‡
Data are means 6 standard deviation and data in parentheses are the range.
§
Comparison according to each TNM stage.
||
Comparison according to low (I or II) and high (III or IV) TNM stage.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
of extrathyroidal extension and lymph node metastasis, instead of the four categories (T1 to T4) of the T stage and the three categories (N0, N1a, N1b) of the N stage. Four sets of the multivariate model were constructed with variables that were statistically significant at univariate analysis as follows: the preoperative model A (age, sex, nodule size, and individual US features), the preoperative model B (age, sex, nodule size, and final US assessment), the postoperative model (age, sex, nodule size, and pathologic characteristics), and the combined model (age, sex, US features, nodule size, and pathologic characteristics). The preoperative model A focused on individual US features, and the preoperative model B focused on the final US assessment, which was either benign or malignant appearing. For the combined model, the final US assessment with the highest HR at univariate analysis was representatively selected instead of individual US features to avoid multicollinearity. Because the internal composition was not included in the criteria that determined if a nodule appeared malignant, it was independently analyzed. All analyses were performed with SPSS software (PASW Statistics, version 20; SPSS, Chicago, Ill). Two-sided P values of less than .05 were considered to indicate a statistically significant difference.
patients were finally found to have biochemical and functional persistent disease (elevated Tg levels and lung
uptake at whole-body scintigraphy), and eight patients had biochemical persistent disease (elevated Tg levels in
Table 2 Characteristics of Patients with and Those without Recurrence during Follow-up Characteristic No. of patients Mean age (y)† Mean age of men (y)† Mean age of women (y)† Male sex US characteristics Internal composition Solid ,50% cystic 50% cystic Echogenicity Hyperechogenicity Isoechogenicity Hypoechogenicity Marked hypoechogenicity Margin Well defined Microlobulated Irregular Calcifications None Macrocalcifications Microcalcifications Shape Parallel Nonparallel Final assessment Benign appearing Malignant appearing Size (mm)‡ Multifocality Extrathyroidal extension Central and/or lateral LN metastasis Central LN metastasis Lateral LN metastasis TNM stage I II III IV Radioiodine ablation therapy
Results Of 515 patients, 56 (10.9%) had recurrence. Forty-four patients had local-regional recurrence (lateral lymph node in 32, thyroid bed in seven, both lateral lymph node and thyroid bed in five), eight patients had distant metastasis (lung in three, lung with other sites [adrenal gland, bone, brain, liver] in five), and four patients had both local-regional (lateral lymph node in one, thyroid bed in one, both lateral lymph node and thyroid bed in two) and distant metastasis (lung in three, lung and bone in one). There were 11 (2.1%) patients with persistent disease. Of them, three
Kim et al
Without Recurrence
With Recurrence
459 (90.1) 45.40 6 12.80 (17–80) 46.09 6 11.42 (21–74) 45.28 6 13.03 (17–80) 65 (14.2)
56 (10.9) 49.05 6 15.47 (17–77) 52.44 6 13.35 (25–69) 47.45 6 16.30 (17–77) 18 (32.1)
427 (93.0) 29 (6.3) 3 (0.7)
46 (82.1) 7 (12.5) 3 (5.4)
3 (0.7) 31 (6.8) 403 (87.8) 22 (4.8)
1 (1.8) 5 (8.9) 46 (82.1) 4 (7.1)
173 (37.7) 226 (49.2) 60 (13.1)
19 (33.9) 26 (46.4) 11 (19.6)
177 (54.5) 32 (7.0) 250 (38.6)
7 (80.4) 4 (7.1) 45 (12.5)
290 (63.2) 169 (36.8)
25 (44.6) 31 (55.4)
109 (23.7) 350 (76.3) 18.81 6 7.76 134 (29.2) 314 (68.4) 273 (59.5)
4 (7.1) 52 (92.9) 27.54 6 12.07 17 (30.4) 49 (87.5) 48 (85.7)
262 (57.1) 84 (18.3)
45 (80.4) 29 (51.8)
258 (56.2) 5 (1.1) 154 (33.6) 42 (9.2) 394 (85.8)
24 (42.9) 0 (0.0) 11 (19.6) 21 (37.5) 43 (76.8)
P Value* .094 .047 .432 .001 .005
.338
.400
,.001
.011
.008
,.001 .980 .005 ,.001 ,.001 ,.001 ,.001§/.040||
.113
Note.—Unless otherwise indicated, data are the number of patients and data in parentheses are percentages. LN = lymph node. * Comparison of patients with and those without recurrence. †
Data are means 6 standard deviation and data in parentheses are the range.
‡
Data are means 6 standard deviation.
§
Comparison according to each TNM stage.
||
Comparison according to low (I or II) and high (III or IV) TNM stage.
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
5
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
seven and an elevated anti-Tg antibody level in one). None of the 11 patients had structural persistent disease.
Malignant- versus Benign-appearing PTC Of 515 patients, 402 (78.1%) had malignant-appearing PTC and 113 (21.9%) had benign-appearing PTC. Patients with malignant-appearing PTC had older mean age (P = .043) but sex was not significantly different (P = .282) (Table 1). The mean follow-up interval was 93 months 6 29 (range, 12–137 months). The mean follow-up interval, internal composition, and size were not significantly different between the two groups (P = .072, .100, and .109, respectively). Malignant-appearing PTC was significantly associated with multifocality, extrathyroidal extension, central and/or lateral lymph node metastasis, central lymph node metastasis, and higher (IIIIV) TNM stage (P = .043, .018, .016, .030, and .002, respectively). Radioactive iodine therapy was not performed differently (P = .793). Association between US Features and Recurrence Mean age was not significantly different between patients with and patients without recurrence (P = .094) (Table 2). More recurrence was observed in male patients than in female patients (P , .001). The mean age was significantly older for men with recurrence (P = .047). Malignant-appearing PTCs were more frequently seen in patients with recurrence than in patients without recurrence (P = .008). Tumor size was larger in patients with recurrence (P , .001). Extrathyroidal extension and central and/or lateral lymph node metastasis, central lymph node metastasis, and lateral lymph node metastasis were more common in patients with recurrence (P = .005, , .001, , .001, and ,.001, respectively). TNM stages were higher in patients with recurrence (P , .001). Patients with malignant-appearing PTC had lower 5- and 10-year diseasefree survival rates than patients with benign-appearing PTC, as demonstrated in Figure 2 (5-year disease-free survival rate of 92.6% [95% CI: 92.1%, 6
93.2%] vs 96.4% [95% CI: 94.5%, 98.3%]; 10-year disease-free survival rate of 84.1% [95% CI: 82%, 86.2%] vs 96.4% [95% CI: 94.5%, 98.3%], respectively) (P = .01). Aging, male sex, cystic component, microcalcifications, taller-thanwide shape, malignant-appearing US features, larger nodule size, and presence of extrathyroidal extension and lymph node metastasis were significantly associated with recurrence at the univariate Cox proportional hazard analysis (Table 3). In the preoperative multivariate model A of the Cox proportional hazard analysis (Table 4), microcalcifications (HR, 2.631; 95% CI: 1.263, 5.482; P = .010), taller-thanwide shape (HR, 3.458; 95% CI: 1.747, 6.846; P , .001), and larger nodule size (HR, 1.076; 95% CI: 1.034, 1.099; P , .001) were independently associated with recurrence. In the preoperative multivariate model B of the Cox proportional hazard analysis, malignant-appearing US features (HR, 3.523; 95% CI: 1.263, 9.830; P = .016) and larger nodule size (HR, 1.074; 95%
Kim et al
CI: 1.051, 1.098; P , .001) were independent factors associated with recurrence. In the postoperative multivariate model of the Cox proportional hazard analysis, male sex (HR, 2.048; 95% CI: 1.148, 3.654; P = .015), larger nodule size (HR, 1.068; 95% CI: 1.044, 1.091; P , .001), extrathyroidal extension (HR, 2.640; 95% CI: 1.184, 5.889; P = .018), and lymph node metastasis (HR, 2.727; 95% CI: 1.273, 5.844; P = .010) were independently associated with recurrence. Recurrence significantly increased 1.021 times a year as patients increased in age (HR, 1.021; 95% CI: 1.002, 1.040; P = .032). In the combined multivariate model of the Cox proportional hazard analysis, male sex (HR, 1.990; 95% CI: 1.098, 3.610; P = .023), malignant-appearing US features (HR, 2.828; 95% CI: 1.016, 7.870; P = .047), larger nodule size (HR, 1.067; 95% CI: 1.043, 1.092; P , .001), extrathyroidal extension (HR, 2.590; 95% CI: 1.160, 5.780; P = .020), and lymph node metastasis (HR, 2.511; 95% CI: 1.163, 5.421; P = .019) were independently associated with recurrence. Age
Figure 2
Figure 2: Kaplan-Meier cumulative event curves for recurrence in benignappearing and malignant-appearing PTCs at preoperative US assessment (log-rank statistic, P = .010). radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
Table 3 Univariate Cox Proportional Hazard Analysis for Association with Recurrence Characteristic Demographics Age Male sex US features Internal composition Solid Mixed cystic and solid Echogenicity Hyper-, iso-, or hypoechogenicity Marked hypoechogenicity Margin Well defined Microlobulated or irregular Calcifications Macrocalcifications or none Microcalcifications Shape Wider than tall Taller than wide Final assessment Benign appearing Malignant appearing Pathologic examination Size Multifocality Extrathyroidal extension Lymph node metastasis Radioactive iodine therapy
HR*
P Value
1.022 (1.002, 1.043) 2.579 (1.472, 4.520)
.035 ,.001
1† 2.485 (1.254, 4.924)
.009
1† 1.720 (0.622, 4.758)
.296
1† 1.151 (0.662, 2.002)
.615
1† 2.923 (1.509, 5.659)
.002
1† 2.000 (1.181, 3.387)
.010
1† 3.484 (1.259, 9.640)
.016
1.079 (1.016, 1.101) 1.086 (0.614, 1.919) 3.100 (1.404, 6.844) 3.848 (1.820, 8.135) 0.584 (0.315, 1.090)
,.001 .777 .005 ,.001 .091
* Data in parentheses are 95% CIs. †
Reference standard.
did not meet statistical significance as an association with recurrence (HR, 1.019; 95% CI: 1.000, 1.038; P = .056).
Discussion In the preoperative model B and the combined model, we found that the presence of preoperative malignant-appearing US features was independently associated with recurrence in patients with classic PTC. Among individual US features, taller-than-wide shape and microcalcifications were independently associated with recurrence in the preoperative model A. In a study by Nam et al, PTCs 10 mm or larger with malignantappearing US features were significantly associated with extrathyroidal extension,
lymph node metastasis, and higher TNM stage, but were not associated with recurrence (19). Recurrence might be underreported due to a relatively short follow-up period of a mean 60 months (range, 18–71 months) and a small cohort of 197 patients with PTCs 10 mm or larger (19). Our longer follow-up period of a mean 93 months 6 29 (range, 12– 137 months) and larger cohort of 515 patients give more adequate information to analyze the effect of preoperative US features on recurrence. Our overall recurrence rate was 10.9%, which was not essentially different from the 9.1% (18 of 197 PTCs 10 mm or larger) observed in a prior study (19). Despite the longer follow-up time and exclusion of papillary thyroid microcarcinomas in our study,
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Kim et al
the comparable recurrence rates may be due to inclusion of patients with total thyroidectomy, prophylactic central neck dissection, and meticulous preoperative US examinations. The 10.9% recurrence rate observed in our study was lower than the 10-year recurrence rates of 14%–26% observed in previous studies (2,4–9). This was due to other variable factors that differed, such as age, sex, tumor size, subtype of PTC, and the extent of surgery, such as prophylactic central compartment dissection (2,4,6,8–10,18). We only included patients with classic PTCs larger than 10 mm who underwent total thyroidectomy and prophylactic central lymph node dissection at the time of operation. Although patients with PTC have excellent prognosis, with 93%–95% 10year survival rates (1–4) and 14%–26% 10-year recurrence rates (2,4–9), 7% have aggressive PTC (32). Given the high annual incidence of PTC (13), a major clinical challenge is to distinguish patients with high risk of recurrence or mortality from those without. Prognostic factors other than age are postoperative factors of tumor size, subtype of tumor, lymph node metastasis, and extrathyroidal extension (5– 8,10–17,33,34). Preoperative prognostic factors can help surgeons decide the extent of the initial thyroid surgery and lymph node dissection. Recently, preoperative BRAF mutation analysis has been shown to identify aggressive PTCs with poor prognostic factors and patients at higher risk of recurrence and mortality (35–38). However, the association between the BRAF mutation and recurrence has been under debate. Preoperative US has been routinely used for patients with PTC, and malignant US features indicative of PTC are well established (13,28,39). On the basis of our results, malignant- or benignappearing US features may be used as a useful, risk-stratifying factor to guide preoperative predictions of prognosis in patients with classic PTC. Benign-appearing PTCs may have delayed diagnosis due to a lack of suspicious US features (19); they may have larger size at diagnosis than those with suspicious US features. However, 7
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
Kim et al
Table 4 Multivariate Cox Proportional Hazard Analysis for Association with Recurrence Preoperative Model A Characteristic Demographics Age Male sex US features Internal composition Solid Mixed cystic and solid Calcifications Macrocalcifications or none Microcalcifications Shape Wider than tall Taller than wide Final assessment Benign appearing Malignant appearing Pathologic features Size Extrathyroidal extension Lymph node metastasis
HR*
Preoperative Model B
P Value
HR*
Postoperative Model
P Value
HR*
1.016 (0.993, 1.039) 2.019 (0.989, 4.122)
.175 .054
1.017 (0.998, 1.036) 1.732 (0.955, 3.141)
.083 .070
1† 1.624 (0.627, 4.206)
… .318
1† 1.152 (0.552, 2.403)
… .707
… …
…
…
1†
Combined Model
P Value
1.021 (1.002, 1.040) 2.048 (1.148, 3.654)
.032 .015
HR*
P Value
1.019 (1.000, 1.038) 1.990 (1.098, 3.610)
.056 .023
… …
1† 1.098 (0.528, 2.329)
… .806
…
…
…
…
2.631 (1.263, 5.482)
.010
…
…
…
…
…
…
1† 3.458 (1.747, 6.846)
… ,.001
… …
… …
… …
… …
… …
… …
… …
… …
1† 3.523 (1.263, 9.830)
… .016
… …
… …
1† 2.828 (1.016, 7.870)
… .047
1.076 (1.034, 1.099) … …
,.001 … …
1.074 (1.051, 1.098) … …
,.001 … …
1.068 (1.044, 1.091) 2.640 (1.184, 5.889) 2.727 (1.273, 5.844)
,.001 .018 .010
1.067 (1.043, 1.092) 2.590 (1.160, 5.780) 2.511 (1.163, 5.421)
,.001 .020 .019
Note.—Preoperative model A: age, sex, composition, individual US features, and nodule size. Preoperative model B: age, sex, composition, final US assessment, and nodule size. Postoperative model: age, sex, nodule size, and pathologic characteristics. Combined model: age, sex, composition, final US assessment, nodule size, and pathologic characteristics. * Data in parentheses are 95% CIs. †
Reference standard.
the mean tumor size was not significantly different between benign- and malignant-appearing PTCs. Larger tumor size was independently associated with recurrence in our study, comparable with prior studies (4,6,8–10). Malignant-appearing PTCs were significantly associated with extrathyroidal extension, lymph node metastasis, and higher TNM stage in our study as was also demonstrated in a prior study (19). Extrathyroidal extension and lymph node metastasis were independently associated with recurrence both in the postoperative and combined models, similar to results from many prior studies (2,6,8,10,16,33). After adjustments for extrathyroidal extension and lymph node metastasis, the presence of malignant-appearing US features was still independently associated with recurrence in the combined multivariate model. Preoperative 8
US features could be used as a preoperative prognostic factor for therapeutic management decisions and as a postoperative prognostic factor to tailor the intensity of postoperative surveillance. Patients with malignant-appearing PTC were older than those with benignappearing PTC (P = .043). However, in the combined multivariate model, older age did not reach statistical significance for association with recurrence. The male-to-female ratio was similar between benign- and malignant-appearing PTC, but male sex was independently associated with recurrence in the combined multivariate model. These results are slightly different from those in prior publications that showed that older age and male sex were associated with recurrence (6,8–11,14,15). However, in male patients, old age was significantly associated with recurrence.
Multifocality was not associated with recurrence (P = .980). The significance of multifocality to influence recurrence is uncertain, as some studies have not found an association between multifocality and recurrence (2,6,11), while others have (4,10). The proportion to undergo radioactive iodine therapy was similar between benign- and malignant-appearing PTCs, and radioactive iodine therapy was not associated with recurrence. Although authors of many studies have found a significant reduction in the rate of recurrence with use of radioactive iodine therapy (2,4,6,10,11,33), others have not observed such benefits (7,40–42). The trend appears to be that the lower the risk, the lower the benefit (41). Of note, our study included patients who underwent total thyroidectomy and prophylactic central compartment dissection, and most of the
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
patients (84.9%) had radioactive iodine therapy. Our study showed that classic PTCs with suspicious US features are more likely to have recurrence than those with benign-appearing US features. We expect that our results can be used in the preoperative risk stratification of classic PTC. We do not anticipate our findings being used for PTCs with suspicious US features that need to undergo aggressive treatment such as postoperative radioactive iodine therapy, because postoperative findings such as size, extrathyroidal extension, and lymph node metastasis are currently being used as parameters to consider radioactive iodine therapy. However, our findings can be used for PTCs with benign-appearing US features that do not need to undergo aggressive treatment (eg, omission of postoperative radioactive iodine therapy and less frequency of postoperative surveillance US). To provide more appropriate recommendations, our findings should be validated in a large, multicenter study. We acknowledge several limitations. First, our study was of retrospective design and patients without preoperative US and those without at least 12 months of follow-up at our institution were excluded. US features were retrospectively reviewed by one of two radiologists. Thus, a bias might exist. Second, radiologists who reviewed the static US features might have been biased toward identifying malignant features, because they knew the purpose of the study, even though they were blinded to pathologic and clinical information. Third, we matched the index tumor of US with that of pathologic examination based on location and the size of the nodule. There was a possibility that the index tumors identified at US may not match those identified in the surgical specimens. In conclusion, the presence of malignant-appearing US features of PTC was independently associated with recurrence in patients with classic PTC. Disclosures of Conflicts of Interest: S.Y.K. disclosed no relevant conflicts of interest. J.Y.K. disclosed no relevant conflicts of interest. E.K.K. disclosed no relevant conflicts of inter-
est. J.H.Y. disclosed no relevant conflicts of interest. H.J.M. disclosed no relevant conflicts of interest.
References 1. Mazzaferri EL, Kloos RT. Clinical review 128: current approaches to primary therapy for papillary and follicular thyroid cancer. J Clin Endocrinol Metab 2001;86(4):1447– 1463. 2. Tsang RW, Brierley JD, Simpson WJ, Panzarella T, Gospodarowicz MK, Sutcliffe SB. The effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer 1998;82(2):375–388. 3. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A National Cancer Data Base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985-1995 [see comments]. Cancer 1998;83(12):2638–2648. 4. Mazzaferri EL, Young RL. Papillary thyroid carcinoma: a 10 year follow-up report of the impact of therapy in 576 patients. Am J Med 1981;70(3):511–518. 5. Simon D, Goretzki PE, Witte J, Röher HD. Incidence of regional recurrence guiding radicality in differentiated thyroid carcinoma. World J Surg 1996;20(7):860–866; discussion 866. 6. DeGroot LJ, Kaplan EL, McCormick M, Straus FH. Natural history, treatment, and course of papillary thyroid carcinoma. J Clin Endocrinol Metab 1990;71(2):414–424. 7. Hay ID, Thompson GB, Grant CS, et al. Papillary thyroid carcinoma managed at the Mayo Clinic during six decades (1940-1999): temporal trends in initial therapy and longterm outcome in 2444 consecutively treated patients. World J Surg 2002;26(8):879–885. 8. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutcliffe SB, Panzarella T. Papillary and follicular thyroid cancer: prognostic factors in 1,578 patients. Am J Med 1987;83(3):479–488. 9. Cady B, Sedgwick CE, Meissner WA, Wool MS, Salzman FA, Werber J. Risk factor analysis in differentiated thyroid cancer. Cancer 1979;43(3):810–820. 10. Loh KC, Greenspan FS, Gee L, Miller TR, Yeo PP. Pathological tumor-node-metastasis (pTNM) staging for papillary and follicular thyroid carcinomas: a retrospective analysis of 700 patients. J Clin Endocrinol Metab 1997;82(11):3553–3562. 11. Samaan NA, Schultz PN, Hickey RC, et al. The results of various modalities of treatment of well differentiated thyroid carcinomas: a
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Kim et al
retrospective review of 1599 patients. J Clin Endocrinol Metab 1992;75(3):714–720. 12. Zaydfudim V, Feurer ID, Griffin MR, Phay JE. The impact of lymph node involvement on survival in patients with papillary and follicular thyroid carcinoma. Surgery 2008;144(6):1070–1077; discussion 1077– 1078. 13. American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer, Cooper DS, Doherty GM, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009;19(11):1167–1214. 14. Tubiana M, Schlumberger M, Rougier P, et al. Long-term results and prognostic factors in patients with differentiated thyroid carcinoma. Cancer 1985;55(4):794–804. 15. Coburn MC, Wanebo HJ. Prognostic factors and management considerations in patients with cervical metastases of thyroid cancer. Am J Surg 1992;164(6):671–676. 16. Hughes CJ, Shaha AR, Shah JP, Loree TR. Impact of lymph node metastasis in differentiated carcinoma of the thyroid: a matchedpair analysis. Head Neck 1996;18(2):127– 132. 17. Moreno Egea A, Rodriguez Gonzalez JM, Sola Perez J, Soria Cogollos T, Parrilla Paricio P. Prognostic value of the tall cell variety of papillary cancer of the thyroid. Eur J Surg Oncol 1993;19(6):517–521. 18. Johnson TL, Lloyd RV, Thompson NW, Beierwaltes WH, Sisson JC. Prognostic implications of the tall cell variant of papillary thyroid carcinoma. Am J Surg Pathol 1988; 12(1):22–27. 19. Nam SY, Shin JH, Han BK, et al. Preop erative ultrasonographic features of papillary thyroid carcinoma predict biological behavior. J Clin Endocrinol Metab 2013;98(4):1476–1482. 20. Hay ID, Grant CS, van Heerden JA, Goellner JR, Ebersold JR, Bergstralh EJ. Papillary thyroid microcarcinoma: a study of 535 cases observed in a 50-year period. Surgery 1992;112(6):1139–1146; discussion 1146– 1147. 21. Ito Y, Miyauchi A, Inoue H, et al. An observational trial for papillary thyroid microcarcinoma in Japanese patients. World J Surg 2010;34(1):28–35. 22. Hay ID, Hutchinson ME, Gonzalez-Losada T, et al. Papillary thyroid microcarcinoma: a study of 900 cases observed in a 60-year period. Surgery 2008;144(6):980–987; discussion 987–988.
9
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
23. Roti E, degli Uberti EC, Bondanelli M, Braverman LE. Thyroid papillary microcarcinoma: a descriptive and meta-analysis study. Eur J Endocrinol 2008;159(6):659–673. 24. Mazzaferri EL, Sipos J. Should all patients with subcentimeter thyroid nodules undergo fine-needle aspiration biopsy and preoperative neck ultrasonography to define the extent of tumor invasion? Thyroid 2008;18(6):597–602. 25. Moon HJ, Son E, Kim EK, Yoon JH, Kwak JY. The diagnostic values of ultrasound and ultrasound-guided fine needle aspiration in subcentimeter-sized thyroid nodules. Ann Surg Oncol 2012;19(1):52–59. 26. Kim JY, Kim EK, Lee HS, Kwak JY. Conventional papillary thyroid carcinoma: effects of cystic changes visible on ultrasonography on disease prognosis. Ultrasonography 2014;33(4):291–297. 27. Kim HG, Kim EK, Han KH, Kim H, Kwak JY. Pathologic spectrum of lymphocytic infiltration and recurrence of papillary thyroid carcinoma. Yonsei Med J 2014;55(4):879– 885. 28. Kim EK, Park CS, Chung WY, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR Am J Roentgenol 2002;178(3):687–691. 29. Moon HJ, Kwak JY, Kim MJ, Son EJ, Kim EK. Can vascularity at power Doppler US
10
Kim et al
help predict thyroid malignancy? Radiology 2010;255(1):260–269.
bases, and clinical implications. Endocr Rev 2007;28(7):742–762.
30. Edge S, American Joint Committee on Cancer. AJCC cancer staging manual. New York, NY: Springer, 2010.
37. Kim TH, Park YJ, Lim JA, et al. The association of the BRAF(V600E) mutation with prognostic factors and poor clinical outcome in papillary thyroid cancer: a meta-analysis. Cancer 2012;118(7):1764–1773.
31. Tuttle RM, Tala H, Shah J, et al. Estimating risk of recurrence in differentiated thyroid cancer after total thyroidectomy and radioactive iodine remnant ablation: using response to therapy variables to modify the initial risk estimates predicted by the new American Thyroid Association staging system. Thyroid 2010;20(12):1341–1349. 32. Cappola AR, Mandel SJ. Molecular testing in thyroid cancer: BRAF mutation status and mortality. JAMA 2013;309(14):1529– 1530. 33. Chow SM, Law SC, Chan JK, Au SK, Yau S, Lau WH. Papillary microcarcinoma of the thyroid: prognostic significance of lymph node metastasis and multifocality. Cancer 2003;98(1):31–40. 34. Akslen LA, LiVolsi VA. Prognostic signifi cance of histologic grading compared with subclassification of papillary thyroid carcinoma. Cancer 2000;88(8):1902–1908. 35. Xing M, Alzahrani AS, Carson KA, et al. Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer. JAMA 2013;309(14):1493–1501. 36. Xing M. BRAF mutation in papillary thy roid cancer: pathogenic role, molecular
38. Xing M, Clark D, Guan H, et al. BRAF mutation testing of thyroid fine-needle aspiration biopsy specimens for preoperative risk stratification in papillary thyroid cancer. J Clin Oncol 2009;27(18):2977–2982. 39. Papini E, Guglielmi R, Bianchini A, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 2002;87(5):1941–1946. 40. Sawka AM, Brierley JD, Tsang RW, et al. An updated systematic review and commentary examining the effectiveness of radioactive iodine remnant ablation in well-differentiated thyroid cancer. Endocrinol Metab Clin North Am 2008;37(2):457–480, x. 41. Van Nostrand D. The benefits and risks of I-131 therapy in patients with well-differentiated thyroid cancer. Thyroid 2009; 19(12):1381–1391. 42. Lundgren CI, Hall P, Dickman PW, Zedenius J. Influence of surgical and postoperative treatment on survival in differentiated thyroid cancer. Br J Surg 2007;94(5):571–577.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
Original Research n Ultrasonography
Association of Preoperative US Features and Recurrence in Patients with Classic Papillary Thyroid Carcinoma1 Soo-Yeon Kim, MD Jin Young Kwak, MD, PhD Eun-Kyung Kim, MD, PhD Jung Hyun Yoon, MD Hee Jung Moon, MD, PhD
Purpose:
To investigate whether ultrasonographic (US) features of thyroid nodules are associated with tumor recurrence in classic papillary thyroid carcinoma (PTC).
Materials and Methods:
This retrospective study was approved by the Institutional Review Board, and the need to obtain informed consent was waived. A total of 515 patients (mean age, 45.8 years 6 13.2 [standard deviation]; range, 17–80 years) who underwent total thyroidectomy and central lymph node dissection for classic PTC greater than 10 mm from January 2003 to February 2006 and who were followed up for 12 months or longer were included. Malignant-appearing PTCs were defined as those showing at least one suspicious US feature among marked hypoechogenicity, irregular or microlobulated margin, microcalcifications, and taller-than-wide shape. Benign-appearing PTCs were defined as those without any suspicious US features. Kaplan-Meier cumulative-event curves for recurrence were compared by using the log-rank test. The multivariate Cox proportional hazard regression analysis was used to estimate hazard ratios (HRs) of the malignant-appearing US features for recurrence in the preoperative, postoperative, and combined models.
Results:
Fifty-six (10.9%) of 515 patients had recurrence. Malignant-appearing PTCs had lower 5- and 10-year diseasefree survival rates compared with benign-appearing PTCs (P = .01). In the preoperative model, malignant-appearing US features (HR, 3.523; 95% confidence interval [CI]: 1.263, 9.830) and larger nodule size (HR, 1.074; 95% CI: 1.051, 1.098) were independently associated with recurrence. In the combined model, male sex (HR, 1.990; 95% CI: 1.098, 3.610), malignant-appearing US features (HR, 2.828; 95% CI: 1.016. 7.870), larger nodule size (HR, 1.067; 95% CI: 1.043, 1.092), extrathyroidal extension (HR, 2.590; 95% CI: 1.160, 5.780), and lymph node metastasis (HR, 2.511; 95% CI: 1.163, 5.421) were independently associated with recurrence.
Conclusion:
The presence of malignant-appearing US features was independently associated with recurrence in patients with classic PTC.
1
From the Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Yonsei University, College of Medicine, 50 Yonsei-ro, Seodaemungu, 120-752 Seoul, Korea. Received October 22, 2014; revision requested December 9; revision received January 20, 2015; accepted February 6; final version accepted February 19 Address correspondence to H.J.M. (e-mail: [email protected]).
RSNA, 2015
q
RSNA, 2015
q
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
1
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
P
apillary thyroid carcinoma (PTC) generally displays an indolent course and has favorable prognosis, with overall 10-year survival rates reported to be 93%–95% (1–4). The reported 10-year recurrence rates are about 14%–26% (2,4–9), and the independent risk factors for recurrence are age at diagnosis, tumor size, pathologic subtype, radioactive iodine therapy, extrathyroidal extension, and lymph node metastasis (2,4–18). Recently, in a report of 488 patients with PTC, ultrasonographic (US) features of PTC at the time of diagnosis served as useful prognostic factors, with poor prognostic factors being more associated with malignant-appearing PTC than benign-appearing PTC (19). Malignant-appearing PTC more frequently had extrathyroidal extension, lymph node metastasis, and a higher TNM stage than benign-appearing PTC (19). However, recurrence was not significantly different between the two groups (19). The mean follow-up time was 60 months (range, 18–71 months); thus, the authors thought that recurrence might be underreported because of the relatively short follow-up periods being observed in a certain portion of patients (19). If the presence of malignant US features is an independent prognostic factor for recurrence, preoperative US features could be used to predict prognosis in individual patients with PTC. These US features could be used to decide the extent of initial surgery, postoperative adjuvant therapy, and the intensity of postoperative surveillance.
Advance in Knowledge nn Malignant-appearing US features of classic papillary thyroid carcinoma (PTC) were significantly associated with recurrence in the preoperative model (hazard ratio, 3.523; 95% confidence interval; 1.263, 9.830; P = .016), as well as the combined model of preoperative and postoperative factors (hazard ratio, 2.828; 95% confidence interval; 1.016, 7.870; P = .047). 2
Classic PTCs 10 mm or smaller have excellent prognosis (19–23). According to a meta-analysis by Roti et al (23), the combination of results from different studies revealed a recurrence rate of 2.4%. PTCs 10 mm or smaller show a higher tendency for false-positive features at US or inadequate cytologic findings at US-guided fine-needle aspiration compared with those larger than 10 mm (24,25). Therefore, we investigated whether US features of thyroid nodules are associated with tumor recurrence in classic PTC larger than 10 mm.
Materials and Methods Patients The Institutional Review Board approved this retrospective study, and informed consent was waived. From January 2003 to February 2006, 2111 consecutive patients underwent thyroid surgery. A total of 1596 patients were excluded: (a) 1133 patients with PTC 10 mm or less, (b) 174 patients with subtotal thyroidectomy or hemithyroidectomy, (c) 150 patients with no available preoperative US images, (d) 49 patients with PTC other than the classic subtype or no PTC, (e) 46 patients with completion thyroidectomy, (f) 30 patients with malignancy other than thyroid cancer, and (g) 14 patients with a follow-up period of less than 12 months after surgery (Fig 1). We included 515 patients who underwent total thyroidectomy for classic PTC larger than 10 mm and had a follow-up period at least 12 months after surgery. Prophylactic central lymph node dissection was routinely performed at the time of total thyroidectomy for all included patients. The mean age of the 515 patients (432 women [mean age, 45.5 years 6 5.5 [standard deviation]; range, 17–80 years] and 83 men [mean age, 47.5 years 6 12.0; range, 21–74 years]) Implication for Patient Care nn The presence of malignantappearing US features of classic PTC is strongly associated with recurrence.
Kim et al
was 45.8 years 6 13.2 (range, 17–80 years). Among the 515 patients, 191 and 94 were previously reported in two published articles, respectively (26,27). However, the association between preoperative US features and recurrence of classic PTC was the primary concern of the present study, and none of these concepts were documented in the previous studies (26,27).
Image Analysis From January 2003 to February 2006, preoperative thyroid US was performed by one of six radiologists, with 1–11 years of experience in thyroid imaging, using 5–15-MHz linear array transducers (HDI 5000, Philips Medical Systems, Bothell, Wash; iU22, Philips Medical Systems; Acuson Sequoia 512 or 513, Siemens Medical Solutions, Mountain View, Calif). When we perform thyroid US in daily practice, we differentiate suspicious nodules from probably benign nodules on the basis of published criteria, such as marked hypoechogenicity, microlobulated or irregular margins, microcalcifications, and taller-than-wide shape (28,29). We report the size and location of a thyroid nodule—whether a nodule is located on (a) the right or left thyroid lobe and on (b) the upper, middle, or lower part of a thyroid lobe in the
Published online before print 10.1148/radiol.2015142470 Content codes: Radiology 2015; 000:1–10 Abbreviations: CI = confidence interval HR = hazard ratio PTC = papillary thyroid carcinoma Tg = thyroglobulin Author contributions: Guarantors of integrity of entire study, J.Y.K., H.J.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, S.Y.K., H.J.M.; clinical studies, J.Y.K., E.K.K., H.J.M.; experimental studies, H.J.M.; statistical analysis, S.Y.K., H.J.M.; and manuscript editing, E.K.K., J.H.Y., H.J.M. Conflicts of interest are listed at the end of this article.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
lymph nodes at level 2, 3, 4, and anterior 5. Therapeutic lateral compartment dissection was performed in 113 patients in our study. Tumor size, multifocality of the tumor, extrathyroidal extension, and central and/or lateral lymph node metastasis status were reassessed and recorded from original pathologic reports without researchers having knowledge of the US features. Index tumors were re-evaluated for TNM staging according to the seventh edition of the American Joint Committee on Cancer staging system and the International Union against Cancer pathologic TNM classification criteria on the basis of their pathologic reports (30).
Figure 1
Figure 1: Flowchart of the study population.
original radiologic report. Also, when we perform fine-needle aspiration, we obtain an image of the needle located within the index nodule. We match the index tumor of US with that of pathologic examination based on location and size. For this study, all US images of index tumors, which were routinely evaluated in both transverse and longitudinal planes and were scanned in both planes, were retrospectively reviewed by one of two radiologists (J.Y.K. for studies performed between January 2003 and August 2004 and H.J.M. for studies performed between September 2004 and February 2006, with 15 and 12 years of experience in thyroid imaging, respectively). There were no special intentions to segregate reviewer cohorts. The US images obtained between January 2003 to August 2004 were previously reviewed by J.Y.K. for the other studies (26,27), and those obtained between September 2004 and February 2006 were reviewed by H.J.M. for this study. The US images were reviewed according to the following five criteria: (a) internal composition (completely solid, or cystic portion , 50% and cystic portion 50% in mixed cystic and solid nodules), (b) echogenicity (hyper-, iso-, or hypoechogenicity, compared with the echogenicity of the underlying thyroid parenchyma or marked hypoechogenicity compared with adjacent strap muscle), (c) margin (well defined, microlobulated, or
Kim et al
irregular), (d) calcifications (no calcification, macrocalcification, or microcalcification [includes nodules with both micro- and macrocalcifications]), and (e) shape (wider than tall, greater in the transverse dimension than the anteroposterior dimension or taller than wide, greater in the anteroposterior dimension than the transverse dimension). The radiologists were blinded to information about the presence of recurrence in patients at the time of US review. Malignant-appearing PTCs were defined as those showing at least one suspicious US feature among marked hypoechogenicity, irregular or microlobulated margins, microcalcifications, and taller-than-wide shape, based on published criteria (28,29). Benignappearing PTCs were defined as those without any suspicious US features.
Surgery and Pathologic Diagnosis In our institution, total thyroidectomy is performed in patients aged 45 years or older with extrathyroidal extension, multifocality, or lymph node metastasis at preoperative or intraoperative examination. Prophylactic central compartment lymph node dissection is routinely performed in patients with total thyroidectomy. Lateral compartment lymph node dissection is performed only when lymph node metastasis is diagnosed at preoperative US-guided fine-needle aspiration or on an intraoperative frozen section. Lateral compartment lymph node dissection includes
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Postoperative Follow-up and Recurrence All patients underwent thyroid-stimulating hormone–suppressive therapy with levothyroxine after surgery. Radioactive iodine therapy was performed in patients with extrathyroidal extension or lymph node metastasis at histopathologic analysis. Four hundred thirtyseven (84.9%) patients underwent radioactive iodine therapy after surgery with 30 mCi (1110 Bq) administered in 420 patients and 50–200 mCi (1850– 7400 Bq) administered in 17 patients. Patients who underwent surgery were followed up every 6 months for the first 3 years and at 12-month intervals thereafter. The routine follow-up protocol consisted of a clinical examination every 6 months; the measurement of serum thyroid-stimulating hormone, free thyroxine, thyroglobulin (Tg), and anti-Tg antibody; a chest x-ray; and a neck US performed every 12 months. Other imaging examinations, such as iodine 131 whole-body scintigraphy, chest computed tomography (CT), or fluorodeoxyglucose positron emission tomography (PET)/CT, were performed in selected cases (eg, detectable or persistent serum Tg or anti-Tg antibody without evidence of recurrence on chest x-rays or neck US scans). No clinical evidence of recurrence was defined as no biochemical (suppressed Tg , 1 ng/mL, stimulated Tg , 2 ng/mL, no detectable anti-Tg antibody), no structural 3
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
(no cytology-proven disease and no morphologic evidence of disease at cross-sectional imaging), and no functional evidence of disease (no suspicious findings at whole-body scintigraphy or PET/CT) (31). Recurrence was defined as new biochemical, structural, or functional evidence of disease that was detected following any period of no evidence of disease (31). Persistent disease was defined as biochemical, structural, or functional evidence of disease that was persistently present without a period of any evidence of disease since the initial surgery.
Data and Statistical Analysis The group with recurrence included patients with recurrence according to the definitions described earlier, and the group without recurrence included patients without recurrence as well as those with persistent disease.
Disease-free survival was defined as the time interval (in months) between surgery and recurrence for the group with recurrence and the time interval between surgery and the latest clinical follow-up for the group without recurrence. We used either the x2 test or the Fisher exact test for categorical variables and the independent t test for continuous variables to compare the differences in demographic, US, and pathologic characteristics between patients with and patients without malignant-appearing US features and between patients with and patients without recurrence. Kaplan-Meier cumulative-event curves were constructed for recurrence in the two groups with benign- or malignantappearing PTC, and the 5-year and 10-year disease-free survival curves were compared with 95% confidence
Kim et al
intervals (CIs) by using the log-rank test. The univariate and multivariate Cox proportional hazard regression models were used to estimate hazard ratios (HRs) with 95% CIs for recurrence according to patient characteristics. Considering the small numbers of recurrence events, all independent variables were classified into binary categories. US features were divided into nonsuspicious and suspicious findings as follows: echogenicity (hyper-, iso-, or hypo- vs marked hypoechogenicity), margin (well defined vs microlobulated or irregular), calcifications (none or macro- vs microcalcification), shape (parallel vs nonparallel), and final assessment (benign vs malignant appearing). Internal composition was divided into completely solid or mixed cystic and solid. Pathologic characteristics were categorized according to the presence
Table 1 Characteristics of All Patients and Patients with Benign- or Malignant-appearing PTC Characteristic No. of patients Mean follow-up (mo)† Mean age (y)‡ Male sex Internal composition at US Solid ,50% cystic 50% cystic Size (mm)† Multifocality Extrathyroidal extension Central and/or lateral lymph node metastasis Central lymph node metastasis Lateral lymph node metastasis TNM stage I II III IV Radioiodine ablation therapy
All Patients
Benign- appearing PTC
Malignant- appearing PTC
515 93 6 29 45.80 6 13.15 (17–80) 83 (16.1)
113 (21.9) 87 6 26 43.58 6 13.46 (17–78) 14 (12.4)
402 (78.1) 94 6 29 46.42 6 13.01 (17–80) 69 (17.2)
473 (91.8) 36 (7.0) 6 (1.2) 19.76 6 8.76 151 (29.3) 363 (70.5) 321 (62.3)
99 (87.6) 13 (11.5) 1 (0.9) 20.92 6 9.07 24 (21.2) 69 (61.1) 59 (52.2)
374 (93.0) 23 (5.7) 5 (1.2) 19.43 6 8.65 127 (31.6) 294 (73.1) 262 (65.2)
307 (59.6) 113 (21.9)
57 (50.4) 18 (15.9)
250 (62.2) 95 (23.6)
282 (54.8) 5 (1.0) 165 (32.0) 63 (12.2) 437 (84.9)
75 (66.4) 3 (2.7) 25 (22.1) 10 (8.8) 95 (84.1)
207 (51.5) 2 (0.5) 140 (34.8) 53 (13.2) 342 (85.1)
P Value* .072 .043 .282 .100
.109 .043 .018 .016 .030 .105 .004§/.002||
.793
Note.—Unless otherwise indicated, data are the number of patients and data in parentheses are percentages. * Comparison of patients with and those without malignant-appearing PTCs.
4
†
Data are means 6 standard deviation.
‡
Data are means 6 standard deviation and data in parentheses are the range.
§
Comparison according to each TNM stage.
||
Comparison according to low (I or II) and high (III or IV) TNM stage.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
of extrathyroidal extension and lymph node metastasis, instead of the four categories (T1 to T4) of the T stage and the three categories (N0, N1a, N1b) of the N stage. Four sets of the multivariate model were constructed with variables that were statistically significant at univariate analysis as follows: the preoperative model A (age, sex, nodule size, and individual US features), the preoperative model B (age, sex, nodule size, and final US assessment), the postoperative model (age, sex, nodule size, and pathologic characteristics), and the combined model (age, sex, US features, nodule size, and pathologic characteristics). The preoperative model A focused on individual US features, and the preoperative model B focused on the final US assessment, which was either benign or malignant appearing. For the combined model, the final US assessment with the highest HR at univariate analysis was representatively selected instead of individual US features to avoid multicollinearity. Because the internal composition was not included in the criteria that determined if a nodule appeared malignant, it was independently analyzed. All analyses were performed with SPSS software (PASW Statistics, version 20; SPSS, Chicago, Ill). Two-sided P values of less than .05 were considered to indicate a statistically significant difference.
patients were finally found to have biochemical and functional persistent disease (elevated Tg levels and lung
uptake at whole-body scintigraphy), and eight patients had biochemical persistent disease (elevated Tg levels in
Table 2 Characteristics of Patients with and Those without Recurrence during Follow-up Characteristic No. of patients Mean age (y)† Mean age of men (y)† Mean age of women (y)† Male sex US characteristics Internal composition Solid ,50% cystic 50% cystic Echogenicity Hyperechogenicity Isoechogenicity Hypoechogenicity Marked hypoechogenicity Margin Well defined Microlobulated Irregular Calcifications None Macrocalcifications Microcalcifications Shape Parallel Nonparallel Final assessment Benign appearing Malignant appearing Size (mm)‡ Multifocality Extrathyroidal extension Central and/or lateral LN metastasis Central LN metastasis Lateral LN metastasis TNM stage I II III IV Radioiodine ablation therapy
Results Of 515 patients, 56 (10.9%) had recurrence. Forty-four patients had local-regional recurrence (lateral lymph node in 32, thyroid bed in seven, both lateral lymph node and thyroid bed in five), eight patients had distant metastasis (lung in three, lung with other sites [adrenal gland, bone, brain, liver] in five), and four patients had both local-regional (lateral lymph node in one, thyroid bed in one, both lateral lymph node and thyroid bed in two) and distant metastasis (lung in three, lung and bone in one). There were 11 (2.1%) patients with persistent disease. Of them, three
Kim et al
Without Recurrence
With Recurrence
459 (90.1) 45.40 6 12.80 (17–80) 46.09 6 11.42 (21–74) 45.28 6 13.03 (17–80) 65 (14.2)
56 (10.9) 49.05 6 15.47 (17–77) 52.44 6 13.35 (25–69) 47.45 6 16.30 (17–77) 18 (32.1)
427 (93.0) 29 (6.3) 3 (0.7)
46 (82.1) 7 (12.5) 3 (5.4)
3 (0.7) 31 (6.8) 403 (87.8) 22 (4.8)
1 (1.8) 5 (8.9) 46 (82.1) 4 (7.1)
173 (37.7) 226 (49.2) 60 (13.1)
19 (33.9) 26 (46.4) 11 (19.6)
177 (54.5) 32 (7.0) 250 (38.6)
7 (80.4) 4 (7.1) 45 (12.5)
290 (63.2) 169 (36.8)
25 (44.6) 31 (55.4)
109 (23.7) 350 (76.3) 18.81 6 7.76 134 (29.2) 314 (68.4) 273 (59.5)
4 (7.1) 52 (92.9) 27.54 6 12.07 17 (30.4) 49 (87.5) 48 (85.7)
262 (57.1) 84 (18.3)
45 (80.4) 29 (51.8)
258 (56.2) 5 (1.1) 154 (33.6) 42 (9.2) 394 (85.8)
24 (42.9) 0 (0.0) 11 (19.6) 21 (37.5) 43 (76.8)
P Value* .094 .047 .432 .001 .005
.338
.400
,.001
.011
.008
,.001 .980 .005 ,.001 ,.001 ,.001 ,.001§/.040||
.113
Note.—Unless otherwise indicated, data are the number of patients and data in parentheses are percentages. LN = lymph node. * Comparison of patients with and those without recurrence. †
Data are means 6 standard deviation and data in parentheses are the range.
‡
Data are means 6 standard deviation.
§
Comparison according to each TNM stage.
||
Comparison according to low (I or II) and high (III or IV) TNM stage.
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
5
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
seven and an elevated anti-Tg antibody level in one). None of the 11 patients had structural persistent disease.
Malignant- versus Benign-appearing PTC Of 515 patients, 402 (78.1%) had malignant-appearing PTC and 113 (21.9%) had benign-appearing PTC. Patients with malignant-appearing PTC had older mean age (P = .043) but sex was not significantly different (P = .282) (Table 1). The mean follow-up interval was 93 months 6 29 (range, 12–137 months). The mean follow-up interval, internal composition, and size were not significantly different between the two groups (P = .072, .100, and .109, respectively). Malignant-appearing PTC was significantly associated with multifocality, extrathyroidal extension, central and/or lateral lymph node metastasis, central lymph node metastasis, and higher (IIIIV) TNM stage (P = .043, .018, .016, .030, and .002, respectively). Radioactive iodine therapy was not performed differently (P = .793). Association between US Features and Recurrence Mean age was not significantly different between patients with and patients without recurrence (P = .094) (Table 2). More recurrence was observed in male patients than in female patients (P , .001). The mean age was significantly older for men with recurrence (P = .047). Malignant-appearing PTCs were more frequently seen in patients with recurrence than in patients without recurrence (P = .008). Tumor size was larger in patients with recurrence (P , .001). Extrathyroidal extension and central and/or lateral lymph node metastasis, central lymph node metastasis, and lateral lymph node metastasis were more common in patients with recurrence (P = .005, , .001, , .001, and ,.001, respectively). TNM stages were higher in patients with recurrence (P , .001). Patients with malignant-appearing PTC had lower 5- and 10-year diseasefree survival rates than patients with benign-appearing PTC, as demonstrated in Figure 2 (5-year disease-free survival rate of 92.6% [95% CI: 92.1%, 6
93.2%] vs 96.4% [95% CI: 94.5%, 98.3%]; 10-year disease-free survival rate of 84.1% [95% CI: 82%, 86.2%] vs 96.4% [95% CI: 94.5%, 98.3%], respectively) (P = .01). Aging, male sex, cystic component, microcalcifications, taller-thanwide shape, malignant-appearing US features, larger nodule size, and presence of extrathyroidal extension and lymph node metastasis were significantly associated with recurrence at the univariate Cox proportional hazard analysis (Table 3). In the preoperative multivariate model A of the Cox proportional hazard analysis (Table 4), microcalcifications (HR, 2.631; 95% CI: 1.263, 5.482; P = .010), taller-thanwide shape (HR, 3.458; 95% CI: 1.747, 6.846; P , .001), and larger nodule size (HR, 1.076; 95% CI: 1.034, 1.099; P , .001) were independently associated with recurrence. In the preoperative multivariate model B of the Cox proportional hazard analysis, malignant-appearing US features (HR, 3.523; 95% CI: 1.263, 9.830; P = .016) and larger nodule size (HR, 1.074; 95%
Kim et al
CI: 1.051, 1.098; P , .001) were independent factors associated with recurrence. In the postoperative multivariate model of the Cox proportional hazard analysis, male sex (HR, 2.048; 95% CI: 1.148, 3.654; P = .015), larger nodule size (HR, 1.068; 95% CI: 1.044, 1.091; P , .001), extrathyroidal extension (HR, 2.640; 95% CI: 1.184, 5.889; P = .018), and lymph node metastasis (HR, 2.727; 95% CI: 1.273, 5.844; P = .010) were independently associated with recurrence. Recurrence significantly increased 1.021 times a year as patients increased in age (HR, 1.021; 95% CI: 1.002, 1.040; P = .032). In the combined multivariate model of the Cox proportional hazard analysis, male sex (HR, 1.990; 95% CI: 1.098, 3.610; P = .023), malignant-appearing US features (HR, 2.828; 95% CI: 1.016, 7.870; P = .047), larger nodule size (HR, 1.067; 95% CI: 1.043, 1.092; P , .001), extrathyroidal extension (HR, 2.590; 95% CI: 1.160, 5.780; P = .020), and lymph node metastasis (HR, 2.511; 95% CI: 1.163, 5.421; P = .019) were independently associated with recurrence. Age
Figure 2
Figure 2: Kaplan-Meier cumulative event curves for recurrence in benignappearing and malignant-appearing PTCs at preoperative US assessment (log-rank statistic, P = .010). radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
Table 3 Univariate Cox Proportional Hazard Analysis for Association with Recurrence Characteristic Demographics Age Male sex US features Internal composition Solid Mixed cystic and solid Echogenicity Hyper-, iso-, or hypoechogenicity Marked hypoechogenicity Margin Well defined Microlobulated or irregular Calcifications Macrocalcifications or none Microcalcifications Shape Wider than tall Taller than wide Final assessment Benign appearing Malignant appearing Pathologic examination Size Multifocality Extrathyroidal extension Lymph node metastasis Radioactive iodine therapy
HR*
P Value
1.022 (1.002, 1.043) 2.579 (1.472, 4.520)
.035 ,.001
1† 2.485 (1.254, 4.924)
.009
1† 1.720 (0.622, 4.758)
.296
1† 1.151 (0.662, 2.002)
.615
1† 2.923 (1.509, 5.659)
.002
1† 2.000 (1.181, 3.387)
.010
1† 3.484 (1.259, 9.640)
.016
1.079 (1.016, 1.101) 1.086 (0.614, 1.919) 3.100 (1.404, 6.844) 3.848 (1.820, 8.135) 0.584 (0.315, 1.090)
,.001 .777 .005 ,.001 .091
* Data in parentheses are 95% CIs. †
Reference standard.
did not meet statistical significance as an association with recurrence (HR, 1.019; 95% CI: 1.000, 1.038; P = .056).
Discussion In the preoperative model B and the combined model, we found that the presence of preoperative malignant-appearing US features was independently associated with recurrence in patients with classic PTC. Among individual US features, taller-than-wide shape and microcalcifications were independently associated with recurrence in the preoperative model A. In a study by Nam et al, PTCs 10 mm or larger with malignantappearing US features were significantly associated with extrathyroidal extension,
lymph node metastasis, and higher TNM stage, but were not associated with recurrence (19). Recurrence might be underreported due to a relatively short follow-up period of a mean 60 months (range, 18–71 months) and a small cohort of 197 patients with PTCs 10 mm or larger (19). Our longer follow-up period of a mean 93 months 6 29 (range, 12– 137 months) and larger cohort of 515 patients give more adequate information to analyze the effect of preoperative US features on recurrence. Our overall recurrence rate was 10.9%, which was not essentially different from the 9.1% (18 of 197 PTCs 10 mm or larger) observed in a prior study (19). Despite the longer follow-up time and exclusion of papillary thyroid microcarcinomas in our study,
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Kim et al
the comparable recurrence rates may be due to inclusion of patients with total thyroidectomy, prophylactic central neck dissection, and meticulous preoperative US examinations. The 10.9% recurrence rate observed in our study was lower than the 10-year recurrence rates of 14%–26% observed in previous studies (2,4–9). This was due to other variable factors that differed, such as age, sex, tumor size, subtype of PTC, and the extent of surgery, such as prophylactic central compartment dissection (2,4,6,8–10,18). We only included patients with classic PTCs larger than 10 mm who underwent total thyroidectomy and prophylactic central lymph node dissection at the time of operation. Although patients with PTC have excellent prognosis, with 93%–95% 10year survival rates (1–4) and 14%–26% 10-year recurrence rates (2,4–9), 7% have aggressive PTC (32). Given the high annual incidence of PTC (13), a major clinical challenge is to distinguish patients with high risk of recurrence or mortality from those without. Prognostic factors other than age are postoperative factors of tumor size, subtype of tumor, lymph node metastasis, and extrathyroidal extension (5– 8,10–17,33,34). Preoperative prognostic factors can help surgeons decide the extent of the initial thyroid surgery and lymph node dissection. Recently, preoperative BRAF mutation analysis has been shown to identify aggressive PTCs with poor prognostic factors and patients at higher risk of recurrence and mortality (35–38). However, the association between the BRAF mutation and recurrence has been under debate. Preoperative US has been routinely used for patients with PTC, and malignant US features indicative of PTC are well established (13,28,39). On the basis of our results, malignant- or benignappearing US features may be used as a useful, risk-stratifying factor to guide preoperative predictions of prognosis in patients with classic PTC. Benign-appearing PTCs may have delayed diagnosis due to a lack of suspicious US features (19); they may have larger size at diagnosis than those with suspicious US features. However, 7
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
Kim et al
Table 4 Multivariate Cox Proportional Hazard Analysis for Association with Recurrence Preoperative Model A Characteristic Demographics Age Male sex US features Internal composition Solid Mixed cystic and solid Calcifications Macrocalcifications or none Microcalcifications Shape Wider than tall Taller than wide Final assessment Benign appearing Malignant appearing Pathologic features Size Extrathyroidal extension Lymph node metastasis
HR*
Preoperative Model B
P Value
HR*
Postoperative Model
P Value
HR*
1.016 (0.993, 1.039) 2.019 (0.989, 4.122)
.175 .054
1.017 (0.998, 1.036) 1.732 (0.955, 3.141)
.083 .070
1† 1.624 (0.627, 4.206)
… .318
1† 1.152 (0.552, 2.403)
… .707
… …
…
…
1†
Combined Model
P Value
1.021 (1.002, 1.040) 2.048 (1.148, 3.654)
.032 .015
HR*
P Value
1.019 (1.000, 1.038) 1.990 (1.098, 3.610)
.056 .023
… …
1† 1.098 (0.528, 2.329)
… .806
…
…
…
…
2.631 (1.263, 5.482)
.010
…
…
…
…
…
…
1† 3.458 (1.747, 6.846)
… ,.001
… …
… …
… …
… …
… …
… …
… …
… …
1† 3.523 (1.263, 9.830)
… .016
… …
… …
1† 2.828 (1.016, 7.870)
… .047
1.076 (1.034, 1.099) … …
,.001 … …
1.074 (1.051, 1.098) … …
,.001 … …
1.068 (1.044, 1.091) 2.640 (1.184, 5.889) 2.727 (1.273, 5.844)
,.001 .018 .010
1.067 (1.043, 1.092) 2.590 (1.160, 5.780) 2.511 (1.163, 5.421)
,.001 .020 .019
Note.—Preoperative model A: age, sex, composition, individual US features, and nodule size. Preoperative model B: age, sex, composition, final US assessment, and nodule size. Postoperative model: age, sex, nodule size, and pathologic characteristics. Combined model: age, sex, composition, final US assessment, nodule size, and pathologic characteristics. * Data in parentheses are 95% CIs. †
Reference standard.
the mean tumor size was not significantly different between benign- and malignant-appearing PTCs. Larger tumor size was independently associated with recurrence in our study, comparable with prior studies (4,6,8–10). Malignant-appearing PTCs were significantly associated with extrathyroidal extension, lymph node metastasis, and higher TNM stage in our study as was also demonstrated in a prior study (19). Extrathyroidal extension and lymph node metastasis were independently associated with recurrence both in the postoperative and combined models, similar to results from many prior studies (2,6,8,10,16,33). After adjustments for extrathyroidal extension and lymph node metastasis, the presence of malignant-appearing US features was still independently associated with recurrence in the combined multivariate model. Preoperative 8
US features could be used as a preoperative prognostic factor for therapeutic management decisions and as a postoperative prognostic factor to tailor the intensity of postoperative surveillance. Patients with malignant-appearing PTC were older than those with benignappearing PTC (P = .043). However, in the combined multivariate model, older age did not reach statistical significance for association with recurrence. The male-to-female ratio was similar between benign- and malignant-appearing PTC, but male sex was independently associated with recurrence in the combined multivariate model. These results are slightly different from those in prior publications that showed that older age and male sex were associated with recurrence (6,8–11,14,15). However, in male patients, old age was significantly associated with recurrence.
Multifocality was not associated with recurrence (P = .980). The significance of multifocality to influence recurrence is uncertain, as some studies have not found an association between multifocality and recurrence (2,6,11), while others have (4,10). The proportion to undergo radioactive iodine therapy was similar between benign- and malignant-appearing PTCs, and radioactive iodine therapy was not associated with recurrence. Although authors of many studies have found a significant reduction in the rate of recurrence with use of radioactive iodine therapy (2,4,6,10,11,33), others have not observed such benefits (7,40–42). The trend appears to be that the lower the risk, the lower the benefit (41). Of note, our study included patients who underwent total thyroidectomy and prophylactic central compartment dissection, and most of the
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
patients (84.9%) had radioactive iodine therapy. Our study showed that classic PTCs with suspicious US features are more likely to have recurrence than those with benign-appearing US features. We expect that our results can be used in the preoperative risk stratification of classic PTC. We do not anticipate our findings being used for PTCs with suspicious US features that need to undergo aggressive treatment such as postoperative radioactive iodine therapy, because postoperative findings such as size, extrathyroidal extension, and lymph node metastasis are currently being used as parameters to consider radioactive iodine therapy. However, our findings can be used for PTCs with benign-appearing US features that do not need to undergo aggressive treatment (eg, omission of postoperative radioactive iodine therapy and less frequency of postoperative surveillance US). To provide more appropriate recommendations, our findings should be validated in a large, multicenter study. We acknowledge several limitations. First, our study was of retrospective design and patients without preoperative US and those without at least 12 months of follow-up at our institution were excluded. US features were retrospectively reviewed by one of two radiologists. Thus, a bias might exist. Second, radiologists who reviewed the static US features might have been biased toward identifying malignant features, because they knew the purpose of the study, even though they were blinded to pathologic and clinical information. Third, we matched the index tumor of US with that of pathologic examination based on location and the size of the nodule. There was a possibility that the index tumors identified at US may not match those identified in the surgical specimens. In conclusion, the presence of malignant-appearing US features of PTC was independently associated with recurrence in patients with classic PTC. Disclosures of Conflicts of Interest: S.Y.K. disclosed no relevant conflicts of interest. J.Y.K. disclosed no relevant conflicts of interest. E.K.K. disclosed no relevant conflicts of inter-
est. J.H.Y. disclosed no relevant conflicts of interest. H.J.M. disclosed no relevant conflicts of interest.
References 1. Mazzaferri EL, Kloos RT. Clinical review 128: current approaches to primary therapy for papillary and follicular thyroid cancer. J Clin Endocrinol Metab 2001;86(4):1447– 1463. 2. Tsang RW, Brierley JD, Simpson WJ, Panzarella T, Gospodarowicz MK, Sutcliffe SB. The effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer 1998;82(2):375–388. 3. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A National Cancer Data Base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985-1995 [see comments]. Cancer 1998;83(12):2638–2648. 4. Mazzaferri EL, Young RL. Papillary thyroid carcinoma: a 10 year follow-up report of the impact of therapy in 576 patients. Am J Med 1981;70(3):511–518. 5. Simon D, Goretzki PE, Witte J, Röher HD. Incidence of regional recurrence guiding radicality in differentiated thyroid carcinoma. World J Surg 1996;20(7):860–866; discussion 866. 6. DeGroot LJ, Kaplan EL, McCormick M, Straus FH. Natural history, treatment, and course of papillary thyroid carcinoma. J Clin Endocrinol Metab 1990;71(2):414–424. 7. Hay ID, Thompson GB, Grant CS, et al. Papillary thyroid carcinoma managed at the Mayo Clinic during six decades (1940-1999): temporal trends in initial therapy and longterm outcome in 2444 consecutively treated patients. World J Surg 2002;26(8):879–885. 8. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutcliffe SB, Panzarella T. Papillary and follicular thyroid cancer: prognostic factors in 1,578 patients. Am J Med 1987;83(3):479–488. 9. Cady B, Sedgwick CE, Meissner WA, Wool MS, Salzman FA, Werber J. Risk factor analysis in differentiated thyroid cancer. Cancer 1979;43(3):810–820. 10. Loh KC, Greenspan FS, Gee L, Miller TR, Yeo PP. Pathological tumor-node-metastasis (pTNM) staging for papillary and follicular thyroid carcinomas: a retrospective analysis of 700 patients. J Clin Endocrinol Metab 1997;82(11):3553–3562. 11. Samaan NA, Schultz PN, Hickey RC, et al. The results of various modalities of treatment of well differentiated thyroid carcinomas: a
Radiology: Volume 000: Number 0— 2015 n radiology.rsna.org
Kim et al
retrospective review of 1599 patients. J Clin Endocrinol Metab 1992;75(3):714–720. 12. Zaydfudim V, Feurer ID, Griffin MR, Phay JE. The impact of lymph node involvement on survival in patients with papillary and follicular thyroid carcinoma. Surgery 2008;144(6):1070–1077; discussion 1077– 1078. 13. American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer, Cooper DS, Doherty GM, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009;19(11):1167–1214. 14. Tubiana M, Schlumberger M, Rougier P, et al. Long-term results and prognostic factors in patients with differentiated thyroid carcinoma. Cancer 1985;55(4):794–804. 15. Coburn MC, Wanebo HJ. Prognostic factors and management considerations in patients with cervical metastases of thyroid cancer. Am J Surg 1992;164(6):671–676. 16. Hughes CJ, Shaha AR, Shah JP, Loree TR. Impact of lymph node metastasis in differentiated carcinoma of the thyroid: a matchedpair analysis. Head Neck 1996;18(2):127– 132. 17. Moreno Egea A, Rodriguez Gonzalez JM, Sola Perez J, Soria Cogollos T, Parrilla Paricio P. Prognostic value of the tall cell variety of papillary cancer of the thyroid. Eur J Surg Oncol 1993;19(6):517–521. 18. Johnson TL, Lloyd RV, Thompson NW, Beierwaltes WH, Sisson JC. Prognostic implications of the tall cell variant of papillary thyroid carcinoma. Am J Surg Pathol 1988; 12(1):22–27. 19. Nam SY, Shin JH, Han BK, et al. Preop erative ultrasonographic features of papillary thyroid carcinoma predict biological behavior. J Clin Endocrinol Metab 2013;98(4):1476–1482. 20. Hay ID, Grant CS, van Heerden JA, Goellner JR, Ebersold JR, Bergstralh EJ. Papillary thyroid microcarcinoma: a study of 535 cases observed in a 50-year period. Surgery 1992;112(6):1139–1146; discussion 1146– 1147. 21. Ito Y, Miyauchi A, Inoue H, et al. An observational trial for papillary thyroid microcarcinoma in Japanese patients. World J Surg 2010;34(1):28–35. 22. Hay ID, Hutchinson ME, Gonzalez-Losada T, et al. Papillary thyroid microcarcinoma: a study of 900 cases observed in a 60-year period. Surgery 2008;144(6):980–987; discussion 987–988.
9
ULTRASONOGRAPHY: Association of Preoperative US Features and Recurrence in Patients with Papillary Thyroid Carcinoma
23. Roti E, degli Uberti EC, Bondanelli M, Braverman LE. Thyroid papillary microcarcinoma: a descriptive and meta-analysis study. Eur J Endocrinol 2008;159(6):659–673. 24. Mazzaferri EL, Sipos J. Should all patients with subcentimeter thyroid nodules undergo fine-needle aspiration biopsy and preoperative neck ultrasonography to define the extent of tumor invasion? Thyroid 2008;18(6):597–602. 25. Moon HJ, Son E, Kim EK, Yoon JH, Kwak JY. The diagnostic values of ultrasound and ultrasound-guided fine needle aspiration in subcentimeter-sized thyroid nodules. Ann Surg Oncol 2012;19(1):52–59. 26. Kim JY, Kim EK, Lee HS, Kwak JY. Conventional papillary thyroid carcinoma: effects of cystic changes visible on ultrasonography on disease prognosis. Ultrasonography 2014;33(4):291–297. 27. Kim HG, Kim EK, Han KH, Kim H, Kwak JY. Pathologic spectrum of lymphocytic infiltration and recurrence of papillary thyroid carcinoma. Yonsei Med J 2014;55(4):879– 885. 28. Kim EK, Park CS, Chung WY, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR Am J Roentgenol 2002;178(3):687–691. 29. Moon HJ, Kwak JY, Kim MJ, Son EJ, Kim EK. Can vascularity at power Doppler US
10
Kim et al
help predict thyroid malignancy? Radiology 2010;255(1):260–269.
bases, and clinical implications. Endocr Rev 2007;28(7):742–762.
30. Edge S, American Joint Committee on Cancer. AJCC cancer staging manual. New York, NY: Springer, 2010.
37. Kim TH, Park YJ, Lim JA, et al. The association of the BRAF(V600E) mutation with prognostic factors and poor clinical outcome in papillary thyroid cancer: a meta-analysis. Cancer 2012;118(7):1764–1773.
31. Tuttle RM, Tala H, Shah J, et al. Estimating risk of recurrence in differentiated thyroid cancer after total thyroidectomy and radioactive iodine remnant ablation: using response to therapy variables to modify the initial risk estimates predicted by the new American Thyroid Association staging system. Thyroid 2010;20(12):1341–1349. 32. Cappola AR, Mandel SJ. Molecular testing in thyroid cancer: BRAF mutation status and mortality. JAMA 2013;309(14):1529– 1530. 33. Chow SM, Law SC, Chan JK, Au SK, Yau S, Lau WH. Papillary microcarcinoma of the thyroid: prognostic significance of lymph node metastasis and multifocality. Cancer 2003;98(1):31–40. 34. Akslen LA, LiVolsi VA. Prognostic signifi cance of histologic grading compared with subclassification of papillary thyroid carcinoma. Cancer 2000;88(8):1902–1908. 35. Xing M, Alzahrani AS, Carson KA, et al. Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer. JAMA 2013;309(14):1493–1501. 36. Xing M. BRAF mutation in papillary thy roid cancer: pathogenic role, molecular
38. Xing M, Clark D, Guan H, et al. BRAF mutation testing of thyroid fine-needle aspiration biopsy specimens for preoperative risk stratification in papillary thyroid cancer. J Clin Oncol 2009;27(18):2977–2982. 39. Papini E, Guglielmi R, Bianchini A, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 2002;87(5):1941–1946. 40. Sawka AM, Brierley JD, Tsang RW, et al. An updated systematic review and commentary examining the effectiveness of radioactive iodine remnant ablation in well-differentiated thyroid cancer. Endocrinol Metab Clin North Am 2008;37(2):457–480, x. 41. Van Nostrand D. The benefits and risks of I-131 therapy in patients with well-differentiated thyroid cancer. Thyroid 2009; 19(12):1381–1391. 42. Lundgren CI, Hall P, Dickman PW, Zedenius J. Influence of surgical and postoperative treatment on survival in differentiated thyroid cancer. Br J Surg 2007;94(5):571–577.
radiology.rsna.org n Radiology: Volume 000: Number 0— 2015
