Treatment Outcomes and Risk Factors for Recurrence After Definitive Surgery of Locally Invasive Well-Differentiated Papillary Thyroid Carcinoma

Abstract

Background:

Papillary thyroid carcinoma (PTC) is generally an indolent tumor that has a favorable prognosis. However, locally invasive PTC can recur after treatment, and its optimal treatment is still controversial. This study aimed to evaluate treatment outcomes and identify risk factors for recurrence and survival in patients with locally invasive PTC.

Materials and Methods:

All consecutive patients who underwent definitive surgery and radioactive iodine therapy for non-distant metastatic invasive PTC were included. Clinical factors, operative and pathological findings, surgical morbidity, and recurrences were recorded. Univariate and multivariate Cox proportional hazard models served to identify factors associated with recurrence-free survival (RFS) and overall survival.

Results:

Of the 96 patients, 74 (77%), 52 (55%), 4 (4%), and 14 (15%) had invasion to the recurrent laryngeal nerve (RLN), trachea, larynx, and esophagus, respectively; 39 (41%) had preoperative vocal cord paresis or paralysis; and 24 (25%) developed recurrence during follow-up (median 77 months). The patients with single and multiple organ invasion did not differ significantly in terms of recurrence-free or overall survival (p > 0.05). The patients with and without recurrences did not differ in terms of surgical extent and involving nerve preservation. Multivariate analysis showed that high (≥1 ng/mL) post-ablation stimulated serum thyroglobulin concentration was an independent predictor of poor RFS (p = 0.013).

Conclusion:

Disease extent, surgical extent, and involving nerve preservation did not associate with recurrence or overall survival outcomes. The post-ablation stimulated thyroglobulin level may be an independent predictor for recurrence. Careful follow-up of patients with this risk factor is recommended.

Introduction

Papillary thyroid carcinoma (PTC) is the most common type of thyroid cancer. It is generally an indolent tumor that is associated with a low rate of cancer-specific mortality (1,2). However, macroscopic extrathyroidal extension (ETE; i.e., local invasion to the adjacent tissue) or nodal metastasis are the main risk factors for recurrence, distant metastasis, and mortality from PTC (3,4). Up to 16% of all patients with PTC have locally advanced PTC (5 –11).

According to the American Joint Committee on Cancer (AJCC) Cancer Staging Manual, T4a stage PTC is defined as a tumor of any size that extends beyond the thyroid capsule to invade the subcutaneous soft tissues, larynx, trachea, esophagus, or recurrent laryngeal nerve (RLN) (12). Thus, T4a tumors can vary markedly in terms of invasion extent and the organ(s) being invaded. While the optimal primary treatment of invasive PTC is complete surgical extirpation, it is not always achievable for T4a tumors because the RLN, larynx, trachea, and esophagus are critical structures whose resection can result in unwanted postoperative complications. The decision not to extirpate the tumor completely can thus leave patients with microscopic and macroscopic residual disease. In such high-risk cases, postoperative radioactive iodine (RAI) ablation with thyrotropin (TSH) suppression may be an effective way to decrease the risk of recurrence and improve survival (13,14). Nevertheless, at present, the optimal surgical approach for T4a tumors remains unclear. There is one study that provides guidance with regard to T4a tumors that involve the RLN, which occurs often. The authors of that study showed that the decision to shave or resect a tumor-involved RLN depends on the presence of preoperative vocal cord paralysis (VCP) (14). However, the optimal surgical approach for tumors that invade the larynx, trachea, and esophagus has not yet been determined.

The standard operative procedures at the authors' tertiary referral center involve intraoperative determination of the extent of infiltration. Superficial invasion of the larynx, trachea, and esophagus (invasion up to the inner cartilage) is then treated by shaving, while intraluminal invasion (invasion of the lining) is treated with wedge resection of the involved organs and reconstruction or anastomosis. When the tumors directly invade the RLN, the nerve is shaved unless the RLN is paralyzed preoperatively or extensively infiltrated, in which case the nerve is often sacrificed. To address whether the standard surgical approaches used at the authors' center affect the recurrence-free and overall survival of patients with T4a PTC, the present study was performed. Thus, all patients with T4a PTC with and without recurrence were compared in terms of their survival outcomes. The effect of extensive disease (involving multiple organs) and surgical extent on recurrence was assessed. In addition, multivariate analyses were performed to identify prognostic factors for recurrence.

Materials and Methods

Study patients

The medical database of the authors' department located in a tertiary referral hospital was reviewed to identify all consecutive patients with PTC who were treated with primary curative surgery between 2005 and 2012. Of the 2784 patients with PTC who underwent surgery during the study period, 121 (4.3%) had stage T4a PTC. Of these patients, 25 were excluded because of distant metastasis at presentation (n = 17) and a prior history of treatment for PTC (n = 8). Finally, 96 patients with well-differentiated T4a PTC were included in the analysis. All study patients underwent preoperative high-resolution ultrasonography (US) evaluation of thyroid nodules and potential metastatic lymph nodes (LNs) in the neck and US-guided biopsy for the diagnosis of thyroid gland malignancies and any central or lateral compartment LNs suspicious for metastasis. Preoperative RLN function of each patient was evaluated by laryngoscopic examination. Computed tomography (CT) or magnetic resonance imaging (MRI) evaluation was also performed in patients with apparent invasion to the larynx, trachea, esophagus, or RLN. The primary endpoint of this study was recurrence-free survival (RFS), and secondary endpoints were morbidities and overall survival. The hospital's Institutional Review Board approved this study. The requirement to obtain written informed consent from each patient was waived.

Surgical approach and adjuvant treatment

All study patients underwent total thyroidectomy and concomitant central compartment neck dissection. Sixty-two patients with suspicious lateral neck metastasis also underwent lateral compartment neck dissection involving levels II–V. When the tumors directly invaded the RLN, the tumor tissue was generally meticulously peeled off the nerve under magnification using surgical loupes. However, when the RLN was paralyzed preoperatively or infiltrated extensively by the tumors, and could not be easily shaved off the tumors, the nerve was often sacrificed. The surgical extent in the trachea, larynx, pharynx, and esophagus was determined by the extent of tumor infiltration into these organs. Superficial invasion of the larynx, pharynx, and trachea was defined as invasion up to the inner cartilage, as determined by intraoperative findings. In these cases, shaving was performed. Intraluminal invasion was defined as invasion of the lining of these organs. In these cases, wedge resection of the involved organs and reconstruction or anastomosis was performed. Circumferential sleeve resection and end-to-end anastomosis were performed in cases of massive tumor infiltration into the lumen and circumferential structures.

One to two months after the initial operation, the patients underwent RAI ablation with 150–200 mCi. Four (4%) patients with obvious macroscopic residual tumors underwent postoperative intensity modulated radiotherapy (PORT) with a total dose of 62–70 Gy (15). Just before RAI ablation, the patients were injected with recombinant TSH to achieve a serum TSH level of >70 IU/mL. The serum thyroglobulin (Tg) levels were measured immediately before (unstimulated Tg) and after (stimulated Tg) the TSH injections (16). TSH stimulation was repeated six to nine months after RAI ablation, and the stimulated Tg levels were measured. Three patients with >100 IU/mL serum anti-Tg antibodies were excluded in the analyses of the relationship between serum Tg results and post-treatment outcomes (17). Biochemical remission (BCR) was defined as a stimulated Tg of <1 ng/mL after the initial operation (18).

Postoperative complications

All patients underwent an indirect laryngoscopic examination for the diagnosis of vocal cord paresis and paralysis before and after the operation (19), and the patients were carefully observed for the development of any complications. Direct laryngoscopy was also performed in patients with stage T4 PTC suspicious of intraluminal invasion into the larynx, trachea, or esophagus at the time of operation prior to neck incision. Patients were clinically evaluated for signs and symptoms of hypocalcemia. Hypocalcemia was defined by total calcium concentrations <8.0 mg/dL during the hospital stay or at any time after discharge from the hospital. Patients who developed hypocalcemia were started on oral calcium and vitamin D replacement and intravenous calcium gluconate for significant symptoms. Permanent hypoparathyroidism was diagnosed in patients with low total calcium concentrations that required calcium supplementation for more than six months.

Post-treatment surveillance

Physical examinations were performed regularly, and the serum Tg levels were checked every 6–12 months. Neck ultrasonography was performed in all patients every 12 months. Additional diagnostic imaging studies such as whole body iodine scanning, neck or contrast-enhanced chest CT were performed when there was clinical suspicion of recurrence or distant metastasis, or when the patient had persistently elevated stimulated Tg levels. ¹⁸F-fluoroedeoxyglucose positron emission tomography/CT was performed when serum the Tg was elevated and the radioiodine scan was negative (16,20). Clinical post-treatment recurrence was defined as the reappearance of pathologically proven tumors in the resection bed or neck, and/or the appearance of metastatic lesions in distant sites.

Statistical analyses

Continuous variables are expressed as median and range or mean ± standard deviation, and categorical variables are expressed as numbers and percentages. Patient groups were compared in terms of dichotomous and continuous variables by using the chi-square (or Fisher's exact) test and the Mann–Whitney U-test, respectively. The RFS and overall survival of patient groups were estimated by using the Kaplan–Meier method and compared by using the log-rank test. The multivariate Cox proportional hazards model was used to analyze the relationship between clinicopathological variables and RFS and overall survival. The hazard ratio (HR) and confidence intervals (CI) were calculated. The organ, as relates to invasion, was the RLN, trachea, esophagus, and larynx. Lymph node ratio represents the number of positive LNs excised divided by the number of LNs excised (21). The Spearman rank correlation test was used to assess the correlation between post-ablation stimulated Tg levels and the number of pathologically positive LNs. All tests were two-sided, and p-values of <0.05 were considered to indicate statistical significance. All statistical analyses were performed by using SPSS Statistics for Windows v22.0 (IBM Corp., Armonk, NY).

Results

Patient characteristics

This study included 96 patients (35 men, 36%). Their median age was 55 years (range 19–85 years). The perioperative clinicopathological characteristics of the patients are shown in Table 1. The pathology of all study patients was well-differentiated PTC without tall-cell or other aggressive variants. Five patients had familial cases with first-degree relatives in families containing two or more similarly affected individuals (22).

Table 1.

Clinical Characteristics of the Study Patients (N = 96)

Variable	n or median (range)	%
Sex, male/female	35/61	36/64
Age, ≤45/>45 years	27/69	28/72
Local invasion to the RLN/trachea/larynx/esophagus^a	74/52/4/14	77/55/4/15
Preoperative vocal cord paresis or paralysis, yes/no	39/57	41/59
Treatment
Surgery alone	3	3
Surgery + RAI/RAI + PORT	89/2	93/2
Surgery + PORT	2	2
Postoperative complication
Hypocalcemia (transient/permanent)	37/11	39/12
Vocal cord paralysis (transient/permanent)	8/53	8/55
Dysphagia	3	3
Follow-up information
Median follow-up (range), months	77 (24–121)
Recurrence (local/nodal/distant)^a	5/15/9	5/16/9
Median time to recurrence (range), months	17 (6–102)
Death (index cancer/other causes)	4/5	4/5

Some patients had invasion in more than one organ.

RAI, radioactive iodine (¹³¹I) therapy; RLN, recurrent laryngeal nerve; PORT, postoperative intensity modulated radiotherapy.

The tumor exceeded 2 cm in 46 (48%) patients, the margin was involved in 59 (62%), the tumor was multifocal in 56 (58%), and lymphovascular invasion was present in 18 (19%). Most patients (n = 91, 95%) underwent surgery and RAI. The remaining five patients underwent surgery with PORT (n = 2) or surgery alone (n = 3) because of their poor general condition (n = 3) or refusal to undergo RAI (n = 2). All patients underwent central compartment neck dissection, and 62 (65%) with suspicious lateral neck metastasis also underwent unilateral (n = 41) or bilateral (n = 21) compartment neck dissection. Pathological nodal involvement was found in 81 (84%) patients: there was central and lateral LN involvement in 22 (23%) and 59 (61%) patients, respectively, with extranodal extension in 54 (56%) patients. The median number of LNs that was excised and pathologically positive was 25 (range 5–154) and 4 (range 0–57), respectively. Fifty-one (53%) patients had five or more positive LNs.

In 74 (77%), 52 (55%), 4 (4%), and 14 (15%) patients, the tumor had invaded into the RLN, trachea, larynx, and esophagus, respectively. Some tumors invaded more than one organ. Of the 74 patients with RLN invasion, 26 and 48 underwent shaving and resection, respectively. Of the 52 patients with trachea involvement, 39 and 13 underwent shaving and resection, respectively. Of the four patients with larynx invasion, three and one underwent shaving and resection, respectively. Of the 14 patients with esophagus invasion, 11 and 3 underwent shaving and resection, respectively.

Of the 96 patients, 39 (41%) patients had preoperative partial vocal cord paresis or complete paralysis on preoperative laryngoscopic examination, and intraoperative examination revealed that 72 (75%) had tumor invasion into the RLN. None of the patients had bilateral RLN involvement. After surgery, 8 (8%) and 53 (55%) patients showed transient and permanent VCP, respectively.

The median serum Tg concentration just before ablation was 0.09 ng/mL (range 0.04–159 ng/mL). After TSH stimulation before ablation, it rose to 2.9 ng/mL (range 0.08–730 ng/mL). Six to nine months after ablation, TSH stimulation was repeated: the median stimulated Tg concentration at that point was 1.30 ng/mL (range 0.08–1510 ng/mL). Of the 91 patients who underwent RAI ablation therapy, 22 (24%) and 41 (45%) had BCR (defined as a stimulated Tg level <1 ng/mL) (18) just before and six to nine months after RAI ablation. The post-ablation stimulated Tg concentration correlated significantly and positively with the number of involved LNs (Spearman's rho = 0.35, p < 0.001; Supplementary Fig. S1; Supplementary Data are available online at www.liebertpub.com/thy).

The mean follow-up period was 77 months (range 24–121 months. Of the 96 patients, 24 (25%) developed recurrence. Of these recurrences, 15 and 9 were locoregional and distant metastasis, respectively. Fifteen patients had nodal recurrences in the lateral neck compartment. The median time to recurrence after surgery was 17 months (range 6–102 months). Reoperative surgery was performed in patients with local and/or nodal recurrence, and ¹³¹I therapy (200 mCi) was administered for the patients with distant metastasis. Nine patients died, four due to the index cancer and five due to other causes.

The patients with single (61 patients, 63.5%) and multiple (35 patients, 36.5%) organ invasion did not differ significantly in terms of RFS and overall survival (p > 0.05).

Comparison of patients with and without post-treatment recurrence

When the 24 patients with recurrence were compared to the 72 patients without recurrence in terms of their clinicopathological variables, they were found to be more likely male (p = 0.002) and to have higher unstimulated and stimulated Tg levels at the time of RAI (p = 0.021 and 0.030, respectively), higher post-ablation stimulated Tg levels (p = 0.011), more extensive neck dissection (p = 0.007), a higher N stage (p = 0.007), higher numbers of excised LNs (p = 0.003), and higher numbers of positive LNs (p = 0.001; Table 2). When superficial and extensive invasion did require shaving or resection, respectively, the two groups did not differ in terms of extent of invasion in the RLN, larynx, and esophagus (p = 0.342, 0.750, and 0.725, respectively). However, the recurrence group was significantly more likely to have extensive trachea invasion (p = 0.006). The two groups did not differ in terms of age, LN ratio, presence of extranodal extension, and use of adjuvant RAI treatment (p > 0.05).

Table 2.

Comparison of the Patients With and Without Recurrence in Terms of Their Clinical Characteristics

Variable	No recurrence (n = 72)	Recurrence (n = 24)	p
Male, n (%)	20 (28)	15 (63)	0.002^*
Age, mean ± SD	54 ± 16	50 ± 17	0.276
Stim. Tg level (ng/mL),^a mean ± SD	16.4 ± 49.7	63.1 ± 152.2	0.030^*
Unstim. Tg level (ng/mL),^a mean ± SD	0.7 ± 2.5	10.3 ± 34.1	0.021^*
Post-ablation stim. Tg (ng/mL),^b mean ± SD	5.2 ± 15.3	117.5 ± 339.0	0.011^*
Extent of neck dissection			0.007^*
Central compartment only, n (%)	31 (43)	3 (13)
Central and lateral compartments, n (%)	41 (57)	21 (88)
Management of adjacent organ
RLN invasion^c	56 (78)	18 (75)
Shaving/resection, n (%)	18/38	8/10	0.342
Trachea	39 (54)	13 (54)
Shaving/resection, n (%)	33/6	6/7	0.006^*
Larynx	3 (4)	1 (4)
Shaving/resection, n (%)	2/1	1/0	0.750
Esophagus	9 (13)	5 (21)
Shaving/resection, n (%)	7/2	4/1	0.725
N stage			0.007^*
pN0, n (%)	15 (21)	0 (0)
pN1a, n (%)	19 (26)	3 (13)
pN1b, n (%)	38 (53)	21 (88)
No. of excised LNs, median (range)	26 (0–89)	54 (4–154)	0.003^*
No. of positive LNs, median (range)	5 (0–27)	15 (2–57)	0.001^*
LN ratio, median (range)^d	22.6 (0–100)	29.2 (3–63)	0.130
Extranodal extension, n (%)	37 (51.4)	17 (70.8)	0.096
Use of adjuvant RAI treatment, n (%)	68 (94.4)	23 (95.8)	0.791

Statistically significant values are shown in bold.

The recurrent cases might also include persistent diseases.

The serum Tg levels just before and after stimulation with TSH, which occurred just before RAI ablation.

A second stimulation with TSH was performed six to nine months after RAI ablation, and the serum Tg levels were measured.

Includes the patients with preoperative vocal cord paresis or paralysis (n = 39).

LN ratio = (number of positive lymph nodes excised/number of lymph nodes excised) ×100 (21).

p < 0.05, as determined by using the chi-square (or Fisher's exact) test and the Mann–Whitney U-test.

LN, lymph node; SD, standard deviation; stim., stimulated; Tg, serum thyroglobulin; TSH, thyrotropin; unstim., unstimulated.

Comparison of patients with RLN invasion who did and did not have preoperative VCP

Of the 74 patients with RLN invasion, 39 (52.7%) had preoperative VCP, as shown by a laryngoscopic examination. When the patients with preoperative non-functioning and functioning RLNs were compared, the non-functioning group was found to be more likely to have multifocal tumors (p = 0.031), develop permanent postoperative VCP (p < 0.001), and undergo RLN resection rather than RLN preservation by shaving (p = 0.005; Table 3). The two groups did not differ in terms of recurrence rates (p = 0.187).

Table 3.

Comparison of Patients Who Did and Did Not Have Preoperative Vocal Cord Palsy (N = 74)

Variable	No preoperative VCP (n = 35)	Preoperative VCP (n = 39)	p-Value
Tumor size, cm, mean ± SD	2.36 ± 1.71	2.77 ± 1.18	0.238
Multifocality, n (%)	24 (69)	17 (44)	0.031^*
Central LN involvement, n (%)	29 (83)	26 (67)	0.111
Extranodal extension, n (%)	18 (51)	26 (67)	0.183
Trachea invasion, n (%)	12 (34)	20 (51)	0.141
Larynx invasion, n (%)	2 (6)	2 (5)	0.911
Esophagus invasion, n (%)	5 (14)	7 (18)	0.670
Handling of the RLN, n (%)			0.005^*
RLN section with tumor	17 (49)	31 (79)
RLN preservation by shaving	18 (51)	8 (21)^a
Postoperative VCP, n (%)			<0.001^*
Transient	6 (17)	1 (3)^a
Permanent	15 (43)	38 (97)
Recurrence, n (%)
Total	14 (40)	10 (26)	0.187
Local/nodal/distant	2/11/4	3/4/5
Local	2 (6)	3 (8)	0.735
Of RLN section patients	0/17 (0)	1/31 (3)	0.560
Of RLN shaving patients	2/18 (11)	2/8 (25)	0.821

Statistically significant values are shown in bold.

Patients with partial paresis but not complete paralysis of the unilateral vocal fold on preoperative laryngoscopy (19).

p < 0.05, as determined by using the chi-square (or Fisher's exact) test and the Mann–Whitney U-test.

VCP, vocal cord palsy.

Of the 35 patients with functioning RLNs, 18 (51%) underwent RLN shaving and 17 (49%) underwent RLN section. Of the 18 patients who underwent RLN shaving, six (33%) had transient VCP and VCP recovery within three to nine months after surgery, and 12 (66%) developed permanent VCP.

Of the 48 patients who had to undergo RLN section due to severe tumor invasion, three (6%) simultaneously underwent unilateral arytenoid adduction. Recurrence rates did not differ between the two groups with and without preoperative VCP, and those with and without involving nerve preservation (p > 0.1).

Risk factors for poor RFS

Univariate analysis revealed that male sex, large (≥2 cm) tumor size, lateral nodal involvement, five or more positive LNs, high (≥1 ng/mL) pre-ablation unstimulated Tg levels, and high (≥1 ng/mL) post-ablation stimulated Tg levels associated significantly with a poor RFS (p < 0.05). RLN, trachea, larynx, and esophagus involvement were not risk factors for poor RFS (p > 0.05) (Table 4). Multivariate analysis with these significant factors showed that high post-ablation stimulated Tg levels independently predicted RFS (HR = 6.81 [CI 1.49–31.14]; p = 0.013).

Table 4.

Univariate and Multivariate Analyses of Factors that Associate with Poor Recurrence-Free Survival

	Univariate		Multivariate
Variable	HR [CI]^a	p-Value^d	HR [CI]	p-Value^d
Sex, male	3.08 [1.35–7.05]	0.008^*	1.19 [0.41–3.42]	0.753
Age, ≥45 years	0.78 [0.33–1.81]	0.556
Family history, yes	1.45 [0.20–10.74]	0.719
Tumor size, >2 cm	3.69 [1.46–9.32]	0.006^*	2.48 [0.79–7.79]	0.121
Positive resection margin	2.17 [0.86–5.47]	0.101
Multifocality	2.26 [0.90–5.70]	0.084
Lymphovascular invasion	1.87 [0.74–4.75]	0.189
Lateral nodal involvement	3.42 [1.17–10.02]	0.025^*	0.75 [0.17–3.41]	0.710
Extranodal extension	2.00 [0.83–4.81]	0.124
≥5 involved LNs	3.05 [1.26–7.36]	0.013^*	3.01 [0.97–9.30]	0.056
RLN involvement	0.61 [0.27–1.40]	0.372
Trachea involvement	1.00 [0.45–2.25]	0.998
Larynx involvement	1.54 [0.21–11.50]	0.675
Esophagus involvement	1.84 [0.68–4.95]	0.230
RAI, yes	1.10 [0.15–8.20]	0.927
PORT, yes	2.83 [0.95–8.39]	0.061
Stim. Tg level (ng/mL),^b ≥1 ng/mL	3.38 [0.79–14.50]	0.101
Unstim. Tg level (ng/mL),^b ≥1 ng/mL	6.48 [2.77–15.13]	<0.001^*	2.10 [0.68–6.47]	0.197
Post-ablation stim. Tg (ng/mL),^c ≥1 ng/mL	8.06 [1.86–34.94]	0.001^*	6.81 [1.49–31.14]	0.013^*

Statistically significant values are shown in bold.

Cox proportional hazards model.

The serum Tg levels just before and after stimulation with thyroid-stimulating hormone, which occurred just before RAI ablation.

A second stimulation with thyroid-stimulating hormone was performed six to nine months after RAI ablation, and the serum Tg levels were measured.

The missing values were replaced with multiple imputations using the Markov chain Monte Carlo method.

p < 0.05.

CI, confidence interval; HR, hazard ratio.

Figure 1 shows the Kaplan–Meier estimates of overall, regional, and distant RFS in patients with post-ablation stimulated Tg levels ≥1 ng/mL (high) and <1 ng/mL (low). Compared with the low Tg group, the high Tg group had significantly lower five-year overall RFS (95% vs. 66%; p = 0.001; Fig. 1A), five-year regional RFS (95% vs. 73%; p = 0.003; Fig. 1B), and five-year distant RFS (100% vs. 87%; p = 0.127; Fig. 1C). Of the nine patients with distant recurrence during follow-up, one and eight were from the low and high Tg groups, respectively. A single patient with a low Tg developed lung metastasis 102 months after primary treatment.

FIG. 1.

Kaplan–Meier estimates showing the overall recurrence-free survival (RFS; A), regional RFS (B), and distant RFS (C) of the patients with low (<1 ng/mL) and high (≥1 ng/mL) post-ablation stimulated thyroglobulin (sTg) level. The cutoff value for the sTg level was defined at 1 ng/mL as described previously (18). Log-rank test, p < 0.05.

Risk factors for poor overall survival

At the last follow-up, four patients had died of PTC and five of unrelated causes. The two- and five-year overall survival rates of the 96 patients were 95% and 92%, respectively. Univariate analysis revealed that a large (>2 cm) tumor size, lymphovascular invasion, laryngeal involvement, PORT, and high unstimulated Tg levels before RAI associated significantly with poor overall survival (p < 0.05; Supplementary Table S1). RLN, trachea, and esophagus involvement were not risk factors for poor RFS (p > 0.05). Multivariate analysis showed that PORT performance remained an independent variable for poor overall survival (HR = 7.64 [CI 1.98–29.39]; p = 0.003).

Discussion

The present retrospective cohort study of patients with T4a PTC showed that 25% developed recurrences during the 77-month follow-up period. The patients with single and multiple organ invasion did not differ significantly in terms of recurrence-free or overall survival. Moreover, the patients with and without recurrence did not differ in terms of surgical extent except for the patients with tracheal involvement: extensive tracheal invasion that required resection was observed more frequently in the patients with recurrence. Multivariate analysis found that RLN, trachea, larynx, and esophagus involvement did not predict poor RFS. Laryngeal, but not RLN, trachea, or esophagus, involvement was associated with worse overall survival. Thus, in general, disease extent did not associate with recurrence or overall survival outcomes. Multivariate analysis showed that a high (≥1 ng/mL) post-ablation stimulated serum Tg concentration was an independent predictor of poor RFS.

Complete resection with tumor-free margins is a fundamental goal of PTC treatment, but in T4a PTC, the resection involves adjacent vital organs whose extensive resection can lead to unwanted outcomes. As a result, the optimal surgical procedures for T4a PTC are at present unclear. It was found that when RLN shaving was used, except in cases of extensive infiltration, surgical difficulty, or preoperative RLN paralysis, RLN involvement did not associate significantly with reduced post-treatment recurrence and survival. Two studies from the Mayo Clinic and Japan, which included patients with functioning but tumor-invaded RLNs who underwent surgery and postoperative RAI, also suggested that patients who require nerve scarification for complete tumor extirpation do not differ in terms of survival outcomes from patients in whom shaving the nerve was sufficient (6,9).

The present study also showed that patients with single adjacent organ (mostly RLN) invasion did not differ from those with multiple organ involvement in terms of RFS and overall survival. This suggests that disease extent does not significantly affect survival outcomes. This was in accordance with the previous studies that showed that the preservation of RLN by shaving does not increase recurrence and disease-specific mortality (5,6,9). By contrast, when Sugitani et al. (23) compared T1–3 tumors with locally advanced T4 tumors, massive ETE (invasion to the RLN, trachea, and/or esophagus) was found to be an important prognostic factor for patients with PTC. Similarly, when Hotomi et al. (24) newly classified the extent of PTC extrathyroidal invasion from Ex0 (no invasion) to Ex3 (invasion to the mucosa of the trachea and/or esophagus, or preoperative VCP), they found that the Ex3 patients had significantly shorter times to regional or distant recurrence and of disease-specific survival (p < 0.05). Similarly, a recent study showed that simultaneous involvement of the trachea and esophagus was associated with higher risk of locoregional recurrence, and laryngeal involvement was associated with lower disease-specific survival in 65 patients with well-differentiated thyroid carcinoma involving the trachea (25). The differences between the present and previous reports might be caused by different inclusion criteria and treatments.

In 23–33% of patients with invasion of the RLN, the voice is affected before surgery (26). VCP is very specific for RLN invasion, although direct pressure on the nerve by an adherent tumor without invasion can also lead to VCP (26,27). The shaving procedure can preserve the normal vocal functionality, even in patients with T4a tumors (10,14). By contrast, patients who have significant tumor invasion and must undergo unilateral RLN section inevitably have poor postoperative vocal function. However, the vocal function and quality of life of these patients can be improved by performing the medialization procedure of the vocal cord immediately after resection or soon in the postoperative period (10). In the present study, the RLN was preserved by the shaving procedure in 18/35 (51%) patients with functioning RLNs. Of these, six did not develop permanent VCP after surgery. Moreover, all three patients who underwent RLN section due to severe tumor invasion and who also underwent simultaneous unilateral arytenoid adduction had good vocal function immediately after surgery. In addition, routine intraoperative nerve monitoring might help nerve handling and predict postoperative RLN function in this setting (28,29).

In the present study, as in other studies, the TSH-stimulated Tg levels served as a surrogate marker of BCR. While BCR can be stringently defined as undetectable TSH-stimulated serum Tg levels, it is also commonly defined as a stimulated Tg levels <1 ng/mL (18) or <2 ng/mL (30,31). Al-Saif et al. showed recently that when they used undetectable stimulated Tg levels to define BCR, only 29% of patients who underwent LN reoperation achieved BCR. However, none of the patients who achieved BCR after reoperation developed post-treatment recurrence (30). Similarly, the present study shows that relatively few T4a patients achieved BCR (24% and 45% before and after RAI, respectively); the patients who lacked BCR had a greater risk of post-treatment recurrence (particularly in the regional neck), and those who achieved BCR did not develop recurrence during the five-year follow-up period. It was also found on multivariate analysis that the post-treatment stimulated Tg level was a predictor of poor RFS. Notably, Lang et al. (31) found that a single post-ablation stimulated Tg level ≥2 ng/mL indicates a high possibility of persistent disease after reoperative neck dissection that requires repeat surgery. Similar findings were obtained when using a Tg cutoff value of 2 ng/mL (data not shown) (30,31).

This study suggests that the patients who do not achieve BCR should be closely monitored for recurrences after surgery for locally advanced PTC. The patients with elevated Tg levels should undergo additional imaging tests for the detection of any structural recurrences. Locoregional recurrences may be treated by reoperative surgery and repeated RAI therapy, while patients with metastatic diseases may be selected for RAI or systemic therapy (16). The patients with persistently elevated Tg levels but no proven structural recurrences might receive repeated RAI therapy. Reoperative surgery is associated with an increased risk of VCP, hypoparathyroidism, and morbidity from lateral neck dissection (32). BCR was also examined in patients after reoperative surgery. BCR was inversely correlated with structural re-recurrence, which sometimes required several surgeries (18,30,33). The postoperative or post-ablation stimulated Tg may be an important predictor of BCR and further recurrence(s) after reoperative neck surgery (18,30,31). Thus, the present observations show that the post-ablation stimulated Tg level is not only a marker for predicting the development of locoregional recurrence in cases of reoperation (30,31), but it also serves the same function in patients who undergo primary curative surgery for locally invasive T4a PTC.

Interestingly, the present study shows that the presence of five or more positive LNs is also associated with a significantly lower RFS on univariate analysis; on multivariate analysis, this association was of borderline significance (p = 0.056). A significant positive correlation between the number of positive LNs and the post-ablation stimulated Tg levels was also observed. This suggests that the number of LNs could also be used to predict recurrence.

This study has the inherent limitations of a retrospective study. There may also be potential biases affecting the risk factor analyses and results. However, this is the first study to show the effect of disease extent on treatment outcomes and the prognostic value of post-ablation stimulated serum Tg levels in patients with locally invasive T4a PTC. The results of this study may help to guide clinicians to identify the at-risk patients with post-treatment recurrence.

In conclusion, the data show that tumor extent—namely, the number of invaded adjacent organs and surgical extent—was not associated with post-treatment recurrence and survival outcomes. They also demonstrate that RLN shaving can preserve normal vocal function, although the complete extirpation of locally invading tumors may provide additional oncological safety. Furthermore, none of the preoperative or intraoperative variables could predict postoperative recurrence. However, post-ablation stimulated serum Tg level was an independent predictor of recurrence.

Footnotes

Acknowledgments

This study was supported by a grant (no. 2015-417) from the Asan Institute for Life Science, a grant (no. 2015R1A2A1A15054540) from Basic Science Research Program through the National Research Foundation of Korea (NRF), Ministry of Science, ICT, and a grant (HI14C23050000) from the Korean Health Technology R&D Project, Ministry of Health & Welfare, Seoul, Republic of Korea (J.-L.R.).

Author Disclosure Statement

The authors declare that they have no conflicts of interest.

References

Grebe

, Hay

. 1996. Thyroid cancer nodal metastases: biologic significance and therapeutic considerations. Surg Oncol Clin North Am, 5:43–63.

Xing

, Alzahrani

, Carson

, Viola

, Elisei

, Bendlova

, Yip

, Mian

, Vianello

, Tuttle

, Robenshtok

, Fagin

, Puxeddu

, Fugazzola

, Czarniecka

, Jarzab

, O'Neill

, Sywak

, Lam

, Riesco-Eizaguirre

, Santisteban

, Nakayama

, Tufano

, Pai

, Zeiger

, Westra

, Clark

, Clifton-Bligh

, Sidransky

, Ladenson

, Sykorova

. 2013. Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer. JAMA, 309:1493–1501.

Gilliland

, Hunt

, Morris

, Key

. 1997. Prognostic factors for thyroid carcinoma. A population-based study of 15,698 cases from the Surveillance, Epidemiology and End Results (SEER) program 1973–1991. Cancer, 79:564–573.

Randolph

, Duh

, Heller

, LiVolsi

, Mandel

, Steward

, Tufano

, Tuttle

. 2012. The prognostic significance of nodal metastases from papillary thyroid carcinoma can be stratified based on the size and number of metastatic lymph nodes, as well as the presence of extranodal extension. Thyroid, 22:1144–1152.

McCaffrey

. 2006. Aerodigestive tract invasion by well-differentiated thyroid carcinoma: diagnosis, management, prognosis, and biology. Laryngoscope, 116:1–11.

Nishida

, Nakao

, Hamaji

, Kamiike

, Kurozumi

, Matsuda

. 1997. Preservation of recurrent laryngeal nerve invaded by differentiated thyroid cancer. Ann Surg, 226:85–91.

Chiang

, Lin

, Lee

, Wang

, Tsai

, Wu

, Lu

, Kuo

. 2006. Thyroid tumors with preoperative recurrent laryngeal nerve palsy: clinicopathologic features and treatment outcome. Surgery, 140:413–417.

Kebebew

, Clark

. 2003. Locally advanced differentiated thyroid cancer. Surg Oncol, 12:91–99.

Falk

, McCaffrey

. 1995. Management of the recurrent laryngeal nerve in suspected and proven thyroid cancer. Otolaryngol Head Neck Surg, 113:42–48.

10.

Roh

, Yoon

, Park

. 2009. Recurrent laryngeal nerve paralysis in patients with papillary thyroid carcinomas: evaluation and management of resulting vocal dysfunction. Am J Surg, 197:459–465.

11.

Chan

, Lo

, Lam

, Wan

. 2004. Recurrent laryngeal nerve palsy in well-differentiated thyroid carcinoma: clinicopathologic features and outcome study. World J Surg, 28:1093–1098.

12.

Edge

SB BD

, Compton

, Fritz

, Greene

, Trotti

. 2010. Thyroid AJCC Cancer Staging Manual. Seventh edition. Springer, New York, NY, pp 87–96.

13.

Kebebew

, Clark

. 2000. Differentiated thyroid cancer: “complete” rational approach. World J Surg, 24:942–951.

14.

Lang

, Lo

, Wong

, Wan

. 2013. Should an involved but functioning recurrent laryngeal nerve be shaved or resected in a locally advanced papillary thyroid carcinoma?. Ann Surg Oncol, 20:2951–2957.

15.

Fussey

, Crunkhorn

, Tedla

, Weickert

, Mehanna

. 2015. External beam radiotherapy in differentiated thyroid carcinoma: a systematic review. Head Neck, 2015 Sep 3. DOI: 10.1002/hed.24218. [Epub ahead of print].

16.

Cooper

, Doherty

, Haugen

, Kloos

, Lee

, Mandel

, Mazzaferri

, McIver

, Pacini

, Schlumberger

, Sherman

, Steward

, Tuttle

. 2009. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid, 19:1167–1214.

17.

Kim

, Kim

, Ryu

, Yeo

, Kim

, Hong

, Shong

. 2005. Serum thyroglobulin levels at the time of ¹³¹I remnant ablation just after thyroidectomy are useful for early prediction of clinical recurrence in low-risk patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab, 90:1440–1445.

18.

Yim

, Kim

, Ryu

, Gong

, Hong

, Shong

. 2011. The outcomes of first reoperation for locoregionally recurrent/persistent papillary thyroid carcinoma in patients who initially underwent total thyroidectomy and remnant ablation. J Clin Endocrinol Metab, 96:2049–2056.

19.

American Academy of Otolaryngology—Head and Neck Surgery. Vocal cord paralysis. Patient health information. Available at: www.entnet.org/content/vocal-cord-paralysis (accessed September 25, 2015 ).

20.

Society of Nuclear Medicine and Molecular Imaging. 2013. 18F-fluorodeoxyglucose (FDG) PET and PET/CT practice guidelines in oncology: a summary of the recommendations and practice guidelines of professional groups. Available at: www.snm.org/docs/PET_PROS/OncologyPracticeGuidelineSummary.pdf (accessed September 25, 2015 ).

21.

Schneider

, Mazeh

, Chen

, Sippel

. 2013. Lymph node ratio predicts recurrence in papillary thyroid cancer. Oncologist, 18:157–162.

22.

McDonald

, Driedger

, Garcia

, Van Uum

, Rachinsky

, Chevendra

, Breadner

, Feinn

, Walsh

, Malchoff

. 2011. Familial papillary thyroid carcinoma: a retrospective analysis. J Oncol, 2011:948786.

23.

Sugitani

, Kasai

, Fujimoto

, Yanagisawa

. 2004. A novel classification system for patients with PTC: addition of the new variables of large (3 cm or greater) nodal metastases and reclassification during the follow-up period. Surgery, 135:139–148.

24.

Hotomi

, Sugitani

, Toda

, Kawabata

, Fujimoto

. 2012. A novel definition of extrathyroidal invasion for patients with papillary thyroid carcinoma for predicting prognosis. World J Surg, 36:1231–1240.

25.

Kim

, Jung

, Lee

, Kwon

, Kim

, Sung

, Hun Hah

. 2015. Prognostic factors of locally invasive well-differentiated thyroid carcinoma involving the trachea. Eur Arch Otorhinolaryngol, 2015 Jul 22. [Epub ahead of print].

26.

Randolph

, Kamani

. 2006. The importance of preoperative laryngoscopy in patients undergoing thyroidectomy: voice, vocal cord function, and the preoperative detection of invasive thyroid malignancy. Surgery, 139:357–362.

27.

Chiang

, Wang

, Huang

, Lee

, Kuo

. 2005. Recurrent laryngeal nerve palsy after thyroidectomy with routine identification of the recurrent laryngeal nerve. Surgery, 137:342–347.

28.

Eid

, Miller

, Rowan

, Otto

. 2013. The role of nerve monitoring to predict postoperative recurrent laryngeal nerve function in thyroid and parathyroid surgery. Laryngoscope, 123:2583–2586.

29.

Calo

, Medas

, Erdas

, Pittau

, Demontis

, Pisano

, Nicolosi

. 2014. Role of intraoperative neuromonitoring of recurrent laryngeal nerves in the outcomes of surgery for thyroid cancer. Int J Surg, 12:S213–217.

30.

Al-Saif

, Farrar

, Bloomston

, Porter

, Ringel

, Kloos

. 2010. Long-term efficacy of lymph node reoperation for persistent papillary thyroid cancer. J Clin Endocrinol Metab, 95:2187–2194.

31.

Lang

, Wong

, Wan

. 2013. Postablation stimulated thyroglobulin level is an important predictor of biochemical complete remission after reoperative cervical neck dissection in persistent/recurrent papillary thyroid carcinoma. Ann Surg Oncol, 20:653–659.

32.

Lee

, Roh

, Gong

, Cho

, Choi

, Nam

, Kim

. 2015. Risk factors for re-recurrence after first reoperative surgery for locoregional recurrent/persistent papillary thyroid carcinoma. World J Surg, 39:1943–1950.

33.

Hughes

, Laird

, Miller

, Gauger

, Doherty

. 2012. Reoperative lymph node dissection for recurrent papillary thyroid cancer and effect on serum thyroglobulin. Ann Surg Oncol, 19:2951–2957.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.04 MB

0.00 MB