Clinical and Pathologic Predictors of Lymph Node Metastasis and Recurrence in Papillary Thyroid Microcarcinoma

Abstract

Background:

The treatment for patients with papillary thyroid microcarcinoma (PTMC) is controversial because PTMC is often found incidentally and its prognosis is very good. Lymph node metastasis (LNM) is one of the main predictors of recurrence and survival. This retrospective study aimed to identify clinical and pathologic factors that increase the risk of metastasis or recurrence, in order to isolate clinically unfavorable PTMCs to help guide therapy.

Methods:

Clinical and pathologic data were collected from 273 patients diagnosed with PTMC at The University of Chicago Medical Center between 2000 and 2011. Data points included age, sex, race/ethnicity, tumor size, multifocality, thyroiditis, extrathyroidal extension (ETE), surgical margins, preoperative clinical suspicion of cancer, central/lateral lymph nodes removed and lymph nodes with metastatic carcinoma, treatment, local recurrence, distant recurrence, and survival.

Results:

Multivariate logistic regression showed that age <45 years (odds ratio [OR] = 3.565 [confidence interval (CI) 1.137–11.177]), multifocality (OR = 3.556 [CI 1.066–11.855]), and ETE (OR = 4.622 [CI = 1.068–20.011]) significantly increased the risk of central LNM (CLNM). However, sex, size of tumor, thyroiditis, positive margins, and clinical suspicion were not correlated with an increased risk for CLNM. Multivariate logistic regression showed that only ETE (OR = 16.066 [CI 1.850–139.488]) significantly increased the risk of lateral LNM. In the cohort of 202 patients with follow-up data, only six recurred. Median time to recurrence was approximately 12 months (range 3.5–120 months), and median follow-up was 42 months. No patient had distant metastasis, and no patients died.

Conclusions:

PTMC is an indolent disease, but does pose a risk for LNM and local recurrence. More aggressive treatment or more frequent follow-up could be considered for patients with unfavorable features (age <45 years, multifocality, ETE), especially in the setting of involved lymph nodes at the time of surgical resection, as these patients may be at an increased risk for recurrence.

Introduction

Thyroid cancer affects >60,000 Americans every year, making it the ninth most common cancer in the United States (1). Papillary thyroid carcinoma (PTC) is the most common malignant tumor of the thyroid gland and accounts for approximately 80% of all thyroid cancers diagnosed in the United States (2). The incidence of thyroid cancer has nearly tripled over the past 35 years, and this increase can partly be accounted for by the increased incidence of papillary thyroid microcarcinomas (PTMCs), which are defined as PTC measuring ≤1 cm in greatest dimension (3). These PTMCs are often found incidentally in thyroid glands that were removed for reasons other than cancer, but may also be detected preoperatively on thyroid ultrasound in conjunction with fine-needle aspiration (FNA).

The treatment for patients with PTMC is controversial because the prognosis of PTMC is very good, with 15-year disease-specific survival rates as high as 99% and recurrence rates as low as 5% (4 –6). Some institutions opt for completion thyroidectomy plus lymph node dissection, while others opt for observation in the event of a solitary PTMC found incidentally on lobectomy/thyroidectomy for otherwise benign disease (7). The lack of a standard treatment may leave clinicians uncertain of how to care for patients with PTMC.

Many studies have shown that lymph node metastasis (LNM) in PTMC is one of the main predictors of recurrence, suggesting that patients presenting with LNM could be treated more aggressively (such as with completion thyroidectomy with or without radioactive iodine [RAI] ablation therapy), as LNM has been shown to decrease overall survival (4,8 –11). Some groups have studied clinical and pathologic features predictive of LNM, which may help guide treatment decisions (12 –17). The results of these studies have varied, and further investigation is warranted.

The current study aimed to identify patients diagnosed with PTMC retrospectively and gather clinical and pathologic data on the primary tumors, as well as metastasis, recurrence, death, and treatment. The study also aimed to identify pathologic factors that may predict metastasis or recurrence in order to isolate clinically unfavorable PTMCs, which may help guide the therapeutic decision-making process for physicians and patients.

Materials and Methods

Patient identification

This retrospective study was performed with Institutional Review Board approval. Patients who underwent thyroid surgery (lobectomy or any type of thyroidectomy) between 2000 and 2011 at The University of Chicago Medical Center and who were found to have a PTC ≤1 cm on pathology were identified. Patients were excluded if they had (i) a previous PTC >1 cm diagnosed at this institution or (ii) previous thyroid resection at another institution with any size PTC or an uncertain diagnosis. A total of 273 patients were included.

Data collection

Pathologic data were collected from surgical pathology reports in the electronic medical record. All of the diagnoses were rendered and reported by University of Chicago pathologists. Clinical and follow-up data were collected from the electronic medical record and through The University of Chicago Cancer Registry.

Data points collected included age, race/ethnicity, tumor size, multifocality, thyroiditis, extrathyroidal extension (ETE), surgical margins, preoperative clinical suspicion of cancer, central/lateral lymph nodes removed, central/lateral lymph nodes with metastatic carcinoma, treatment, as well as local recurrence, distant recurrence, and survival. Age was dichotomized at 45 years in accordance with current staging criteria (18). Tumor size was defined as the largest dimension. Multifocality was defined as >1 focus of tumor in total (either in the same lobe or different lobes). Thyroiditis included any type of thyroiditis mentioned in pathology reports, including Hashimoto's thyroiditis or nonspecific lymphocytic thyroiditis. Surgical margins were considered involved (positive) or uninvolved (negative) based on pathology reports. A PTMC was classified as “clinically suspected” if the patient had an FNA prior to surgery read as “PTC” or “suspicious for PTC,” or if a lymph node resected during unrelated neck surgery demonstrated PTC. Patient were considered to have lymph nodes removed if any number of nodes were resected and examined histologically, from single nodes to complete neck dissections. Patients were classified as positive for LNM if there was PTC in the lymph node on histologic examination. Central lymph nodes included any of the following: central compartment, pretracheal, paratracheal, and prelaryngeal/Delphian lymph nodes (or equivalent to pathologic stage N1a). Lateral lymph nodes included any of the following: unilateral, bilateral, or contralateral cervical (Levels I–V) or retropharyngeal or superior mediastinal (Level VII) lymph nodes (or equivalent to pathologic stage N1b). Treatment was divided into “total thyroidectomy” (which included both one-stage total thyroidectomy, as well as lobectomy followed by completion thyroidectomy at a later date), “total thyroidectomy + post-operative RAI ablation therapy” and “non-total thyroidectomy” (which included lobectomies, sub-total thyroidectomies, and nodulectomies). Of note, total thyroidectomy was considered to consist of removal of all visible thyroid tissue with <1 g of tissue remaining. Therefore, near-total thyroidectomies were included in the total thyroidectomy category. Recurrence was considered proven if diagnosed pathologically (biopsy or resection), and was considered highly likely in the setting of suspicious neck imaging and supporting serologic studies. Local recurrence included a recurrence of PTC in the thyroid bed or regional lymph nodes. Distant recurrence included recurrence in non-regional lymph nodes or visceral sites (such as the lung, brain, or bone).

In cases where the parameters in the original pathology report were unclear, the original slides were re-reviewed.

Statistical analysis

Microsoft Excel (Redmond, WA) and Stata v13 (College Station, TX) were used for statistical analysis. A p-value of <0.05 was considered significant. Multivariate logistic regression was used to calculate odds ratios of clinical and pathologic factors predictive of central LNM (CLNM) and lateral LNM (LLNM).

Results

Patient characteristics

The study population included 273 patients. The average age was 49 years (range 15–83 years). The majority of patients were white (85%) and female (85%). The average size of PTMC (greatest dimension) was 0.49 cm (range 0.01–1 cm). Of all 273 patients, 163 (60%) had lymph nodes removed. In the majority of patients (n = 187; 68%), the microcarcinoma was unsuspected prior to surgery. LNM occurred in 19% (n = 18) and 29% (n = 20) of clinically unsuspected and suspected cases, respectively. Table 1 lists the clinical and pathologic characteristics of all patients, as well as the subset with any lymph nodes removed.

Table 1.

Patient Characteristics

		Patients with LNs removed (n = 163)
	All patients (n = 273)	Negative LNs (n = 125; 77%)	Positive LNs (n = 38; 23%)
Age
<45 years	100 (36.6%)	49 (71%)	20 (29%)
≥45 years	173 (63.4%)	76 (80.9%)	18 (19.1%)
Sex
Male	41 (15%)	14 (66.7%)	7 (33.3%)
Female	232 (85%)	111 (78%)	31 (22%)
Race
White	171 (85%)	84 (75%)	28 (25%)
Black	17 (8.5%)	5 (71%)	2 (29%)
Asian	9 (4.5%)	5 (71%)	2 (29%)
Hispanic	4 (2%)	2 (100%)	0 (0%)
Size
<5 mm	121 (44%)	50 (82%)	11 (18%)
≥5 mm	152 (56%)	75 (73.5%)	27 (26.5%)
Multifocality
Absent	161 (59%)	77 (86.5%)	12 (13.5%)
Present	111 (41%)	47 (64%)	26 (36%)
Thyroiditis
Absent	153 (56%)	49 (70%)	21 (30%)
Present	120 (44%)	76 (82%)	17 (18%)
ETE
Absent	245 (90%)	118 (82.5%)	25 (17.5%)
Present	27 (10%)	7 (37%)	12 (63%)
Margins
Negative	254 (93%)	116 (77%)	34 (23%)
Positive	19 (7%)	9 (70%)	4 (30%)
Clinical suspicion
Unsuspected	187 (68%)	75 (81%)	18 (19%)
Suspected	86 (32%)	50 (71%)	20 (29%)
Lymph node status
Nx	110 (40%)
N0	125 (46%)
N1a	20 (7.5%)
N1b	18 (6.5%)
Treatment ^a
Non-total thyroidectomy	36 (18%)	31 (91%)	3 (9%)
Total thyroidectomy	93 (46%)	69 (83%)	14 (17%)
Total thyroidectomy + RAI	73 (36%)	25 (54.5%)	21 (45.5%)

Includes only patients with >3 months of follow-up.

LN, lymph node; ETE, extrathyroidal extension; RAI, radioiodine ablation therapy.

Lymph node dissection

Of all 273 patients, 163 (60%) had lymph nodes removed: 96 had only central lymph nodes removed, 35 had only lateral lymph nodes removed, and 32 had both central and lateral lymph nodes removed. Of these 163 patients, 38 (23%) had LNM. Of these 38 patients with LNM, 20 had CLNM only (53%), 14 had LLNM only (37%), and four had both CLNM and LLNM (11%).

Risk factors for CLNM

Of the 128 patients who had central lymph nodes removed, 24 (19%) had CLNM. Multivariate logistic regression showed that age <45 years (odds ratio [OR] = 3.565 [confidence interval (CI) 1.137–11.177]), multifocality (OR = 3.556 [CI 1.066–11.855]), and ETE (OR = 4.622 [CI 1.068–20.011]) all significantly increased the risk of CLNM (Table 2). However, sex, size of tumor (≥5 mm), thyroiditis, positive margins, and clinical suspicion were not significantly correlated with an increased risk for CLNM (Table 2).

Table 2.

Multivariate Logistic Regression Analysis for Clinical and Pathologic Features Predictive of Lymph Node Metastasis in Papillary Thyroid Microcarcinoma

	CLNM			LLNM			Any LNM
Risk predictor ^a	OR	CI	p-Value	OR	CI	p-Value	OR	CI	p-Value
Age <45 years	3.565	1.137–11.177	0.029	0.412	0.083–2.056	0.280	1.717	0.723–4.076	0.220
Female	3.058	0.303–30.862	0.343	0.277	0.053–1.447	0.128	0.587	0.181–1.909	0.376
Size ≥5 mm	2.579	0.595–11.170	0.205	0.279	0.052–1.508	0.138	1.382	0.463–4.127	0.562
Multifocality (+)	3.556	1.066–11.855	0.039	2.346	0.608–9.050	0.216	3.264	1.387–7.682	0.007
Thyroiditis (+)	0.458	0.145–1.442	0.182	1.231	0.302–5.022	0.772	0.525	0.218–1.264	0.151
ETE (+)	4.622	1.068–20.011	0.041	16.066	1.850–139.488	0.012	5.945	1.810–19.533	0.003
Margins (+)	1.165	0.112–12.131	0.899	0.775	0.099–6.095	0.809	0.620	0.108–3.574	0.593
Clinical suspicion (+)	0.780	0.201–3.026	0.719	0.264	0.042–1.666	0.156	0.835	0.289–2.415	0.739

Statistically significant values are in bold.

Reference groups: age ≥45 years, male, size <5 mm, multifocality (–), thyroiditis (–), ETE (–), margins (–), clinical suspicion (–).

CLNM, central lymph node metastasis; LLNM, lateral lymph node metastasis; LNM, lymph node metastasis; OR, odds ratio; CI, confidence interval.

Size was also dichotomized at each integer, and it was found that CLNM was not predicted by logistic regression for any tumor size (i.e., ≥1, 2, 3, 4, 5, 6, 7, 8, or 9 mm). In other words, at no size cutoff was risk for CLNM significantly increased.

Risk factors for LLNM

Of the 67 patients who had lateral lymph nodes removed, 18 (27%) had LLNM. Multivariate logistic regression showed that only ETE (OR = 16.066 [CI 1.850–139.488]) significantly increased the risk of LLNM, while age (<45 years), sex, size of tumor (≥5 mm), multifocality, thyroiditis, positive margins, and clinical suspicion were not significantly correlated with an increased risk for LLNM (Table 2).

We also dichotomized size at each integer, and found that LLNM was not predicted by logistic regression for any tumor size (i.e., ≥1, 2, 3, 4, 5, 6, 7, 8, or 9 mm). In other words, at no size cutoff was risk for LLNM significantly increased.

An attempt was made to determine if CLNM predicted LLNM, but it was not possible to fit these into the regression model due to limited patients.

Risk factors for any (either central or lateral) LNM

All 163 patients with lymph node dissection were included in a multivariate logistic regression model to determine factors predictive of LNM in any lymph nodes, and it was found that only multifocality (OR = 3.264 [CI 1.387–7.682]; p = 0.007) and ETE (OR = 5.945 [CI 1.810–19.533], p = 0.003) were significant risk factors for having any LNM (Table 2).

Additionally, a receiver operating characteristic curve analysis was done to determine whether tumor size alone (between 0 and 1 cm) could be predictive of metastasis to central or lateral lymph nodes (Fig. 1). The area under the curve was 0.5420, indicating that there is no size threshold for microcarcinomas that can accurately predict metastasis to lymph nodes.

FIG. 1.

Receiver operating characteristic curve of the ability of tumor size to predict the likelihood of metastasis to either central or lateral lymph nodes. Area under the curve = 0.5420.

Age

Risk for CLNM was also analyzed in patients diagnosed at an older age (>60 years). It was found that compared with patients aged <45 years (n = 69), patients aged 45–60 years (n = 66) had a lower odds ratio of CLNM (OR = 0.238 [CI 0.068–0.836]), and patients aged >60 years (n = 28) had no difference in risk (OR = 0.485 [CI 0.080–2.953]).

Treatment

Of all 273 patients, 202 (74%) had at least three months of follow-up, and treatment data were collected for this subset. Of these, most underwent at least total thyroidectomy (n = 166; 82%), and 44% of these patients also received RAI (n = 73). Most (n = 148) underwent a primary total thyroidectomy, and 18 had a lobectomy followed by a completion thyroidectomy at a later date. Thirty-six (18%) underwent non-total thyroidectomy. Seventy-two percent of patients with LLNM (10/14), 85% of patients with CLNM (17/20), 68% of patients with ETE (15/22), and 54% of patients with multifocality (48/89) were treated with total thyroidectomies + RAI.

Recurrence

Of the 202 patients with at least three months of follow-up, the average follow-up time was 49 months (standard deviation ±31 months), and the median follow-up was 42 months (range 3–193 months). Six (3%) had a recurrence. Four were diagnosed prior to surgery. Five were treated with total thyroidectomy + RAI (one with lobectomy followed by completion); one was treated with surgery only. Median time to recurrence was approximately 12 months (range 3.5–120 months). None of the patients with recurrence had distant metastasis; all had locoregional recurrence (Table 3). At diagnosis, all of the patients with a recurrence had LNM plus one additional unfavorable feature: namely, 5/6 patients had multifocal tumors, and 3/5 had ETE. Tumor size for all patients with recurrence was >0.5 cm in greatest dimension. Of the 202 patients with follow-up, there were six deaths, none due to PTMC.

Table 3.

Clinical and Pathologic Features of Patients with Recurrent Papillary Thyroid Microcarcinoma

Patient	Age (years)	Sex	Size of tumor (cm)	Multifocality	Thyroiditis	ETE	Margin	Clinical suspicion	CLNM	LLNM	Treatment	Time to recurrence (months)	Region of recurrence
1	52	M	0.5	+	−	−	−	+	UNK	+	TT	3.5	Central LN (paratracheal)
2	59	F	0.7	−	+	+	+	+	UNK	+	TT + RAI	11	Thyroid bed
3	22	F	0.9	+	−	+	+	+	+	+	TT + RAI	27	Lateral LN (supraclavicular)
4	51	F	0.8	+	+	+	-	+	+	UNK	TT + RAI	13	Lateral LN (internal jugular)
5	63	F	0.7	+	+	−	+	−	+	UNK	Lobectomy + Completion + RAI	8	Lateral LN (Level II)
6	31	F	0.7	+	+	UNK	+	− S/C	+	+	TT + RAI	120	Central LN (paratracheal, suprasternal)

TT, total thyroidectomy; +, present; −, absent; UNK, unknown; S/C, see explanation in results (Detailed analysis of recurrence patients).

Multivariate logistic regression model (to determine independent risk factors predictive of recurrence while controlling for treatment) was unsuccessful due to the small number of patients with recurrence.

Detailed analysis of recurrence patients

Patient 1 was a 52-year-old male who underwent a parathyroidectomy for a parathyroid adenoma. PTC was found in tracheoesophageal groove lymph nodes that appeared abnormal intraoperatively. Two months later, the patient underwent total thyroidectomy with left cervical lymph node dissection, which revealed two 0.5 cm foci of PTC and 1/9 positive lymph nodes. ETE was absent, and margins were uninvolved (stage pT1a(m), N1b). Three and a half months later, he developed an enlarged paratracheal lymph node that was biopsied, shown to be positive for PTC, and subsequently excised. Given the short time frame (3.5 months), this finding may represent recurrent or residual disease. No additional follow-up is available.

Patient 2 was a 59-year-old female who was evaluated for throbbing sensations in her throat, was found to have a nodule on thyroid ultrasound, and diagnosed with PTC on FNA. She underwent total thyroidectomy with right jugular lymph node dissection, which revealed a solitary 0.7 cm focus of PTC and 1/3 positive lymph nodes. Minimal ETE into fibroadipose tissue was present, and margins were involved (pT3, N1b). She received 200 mCi ¹³¹I. Eleven months later, the patient was noted to have biochemical evidence of recurrence on a recombinant human thyrotropin scan (thyrotropin rose from 0.8 to 25 mIU/L, and triiodothyronine rose from 2 to 9 ng/mL), along with a new 0.7 cm focus of suspicious vascular tissue in the thyroid bed on ultrasound. She received an additional 180 mCi ¹³¹I for suspected recurrence. Six months later, she was disease free.

Patient 3 was a 22-year-old female who was diagnosed with PTC at an outside hospital on neck biopsy. She underwent total thyroidectomy with bilateral cervical lymph node dissection, which revealed multifocal PTC (largest 0.9 cm), 11/14 positive central neck lymph nodes, 5/22 positive right neck lymph nodes, and 12 negative left neck lymph nodes. There was also ETE into skeletal muscle and positive margins (stage pT3, N1b). She received 154 mCi ¹³¹I. Twenty-seven months later, she developed an enlarged supraclavicular lymph node that was biopsied, shown to be positive for PTC, and subsequently excised. Over the next four years, she had multiple local recurrences treated surgically, and was disease free at last follow-up seven years after original diagnosis.

Patient 4 was a 51-year-old female with a thyroid nodule found on palpation and diagnosed with PTC on FNA. She underwent total thyroidectomy with central lymph node dissection (CLND), which revealed multifocal PTC (largest 0.8 cm) and 1/3 positive central neck lymph nodes. ETE into skeletal muscle was present, but margins were uninvolved (stage pT3, N1a). She received 152 mCi ¹³¹I. Thirteen months later, she developed a recurrence in the left internal jugular chain (7/26 positive lymph nodes excised). No additional follow-up is available.

Patient 5 was a 63-year-old female with hyperparathyroidism, multinodular goiter, and difficulty swallowing. She underwent parathyroidectomy and left thyroid lobectomy for symptomatic relief, which revealed multifocal PTC (largest 0.7 cm) and 1/6 positive perithyroidal lymph nodes. ETE was absent, but margins were involved (stage pT1a(m), N1a). She then underwent completion thyroidectomy, which revealed additional foci of PTC and one negative lymph node. She received 108 mCi ¹³¹I. Eight months later, she developed a recurrence in the right neck (2/26 positive lymph nodes excised). She received an additional 150 mCi ¹³¹I. Approximately four years after diagnosis, she was disease free.

Patient 6 was a 31-year-old female who was found to have low cervical lymphadenopathy during evaluation for rotator cuff injury. FNA of a cystic lymph node was non-diagnostic. However, metastatic thyroid cancer was clinically favored. She underwent total thyroidectomy with right cervical lymph node dissection, which revealed multifocal PTC (largest 0.7 cm) and at least 4/18 positive lymph nodes. ETE status was unknown (not stated in report, and slides not available for review). Margins were involved (stage pT1a(m), N1b). She received 150 mCi ¹³¹I. Ten years (120 months) later, she developed recurrence in the paratracheal and suprasternal lymph nodes (4/6 positive lymph nodes excised). She received an additional 101 mCi ¹³¹I. Approximately 11 years after initial diagnosis, she was disease free.

Discussion

This study investigated clinical and pathologic features predictive of LNM and recurrence in patients with PTMCs. The study confirms the indolent behavior of PTMC, as only 3% of patients in the cohort had recurrence, and none died due to PTMC. Patients who were <45 years of age were significantly more likely to have CLNM. Tumors with multifocality or ETE on histologic examination were significantly more likely to have CLNM; tumors with ETE were also significantly more likely to have LLNM. PTMCs found incidentally were at equal risk of LNM compared with those diagnosed by a preoperative positive or suspicious FNA.

These findings are somewhat similar to other studies that have examined predictors of CLNM in PTMC. Most, including the present study, found that ETE, multifocality, and younger age (either <50 or <45 years) are predictors of CLNM (Table 4). Male sex and tumor size were not predictive of CLNM in the current study, in contrast to findings by others (8,12,14,19,20). The smallest tumor to metastasize measured 0.2 cm in greatest dimension. In this case, multifocality and ETE were absent. However, the patient was 41 years old. Recent studies have found that patients diagnosed with PTC between age 45 and 60 years of age have a greater relative risk of dying from thyroid cancer compared with patients >60 years at diagnosis (21). An attempt was made to determine if age >60 years predicted increased risk of LNM in PTMC. However, no significant difference was found compared to age <45 years or age 45–60 years. The study did find that compared with patients aged <45 years, patients aged 45–60 years had a lower risk of CLNM.

Table 4.

Concise Literature Review of Factors Predictive for CLNM

Article	Size of study	Age	Sex (male)	Size of tumor	Multifocality	Bilaterality	Thyroiditis	ETE	Positive margins	Clinical suspicion	Subtype (follicular)	LLNM	BRAF mutation
Liu, 2014 (8)	1928	NS	NS	OR = 2.3 (>0.5 cm)	OR = 1.8	—	—	OR = 2.8	—	—	—	—	—
Zhang, 2012 (12)	1066	OR = 1.4 (age ≤45 years)	OR = 1.9	OR = 1.4 (>0.6 cm)	OR = 4.5	NS	—	OR = 4.2	—	—	NS	—	—
Chang, 2015 (28)	613	NS	OR = 2.1	OR = 2.3 (≥0.5 cm)	OR = 1.76	—	NS	OR = 1.49	—	—	—	—	—
Xu, 2014 (29)	507	NS	NS	OR = 0.52 (>0.5 cm)	NS	OR = 0.34	—	OR = 2.5	—	—	—	—	—
Yang, 2014 (14)	291	NS	OR = 2.0	OR = 3.7 (>0.5 cm)	NS	NS	NS	OR = 2.3	—	—	—	OR = 15.1	OR = 2.5
Pisanu, 2015 (30)	219	—	—	OR = 2.6 (>0.8 cm)	OR = 5.9	—	—	OR = 10.3	—	—	—	—	—
Park, 2014 (31)	193	NS	NS	OR = 2.28 (>0.7 cm)	OR = 2.38	—	NS	NS	—	—	—	—	—
Zhang, 2015 (32)	178	NS	OR = 0.13	OR = 6.6 (≥0.6 cm)	NS	NS	NS	OR = 0.18	—	—	—	—	—
Zhou, 2012 (19)	122	OR = 3.4 (age <50 years)	OR = 4.8	OR = 2.9 (≥0.7 cm)	NS	NS	NS	NS	—	—	—	—	—
Xu, 2014 (20)	85	—	—	OR = 3.9 (>0.5 cm)	NS	NS	—	OR = 3.9	—	—	—	—	—
Current study	163	OR = 3.6 (age <45 years)	NS	NS	OR = 3.6	—	NS	OR = 4.6	NS	NS	—	—	—

—, article did not study factor; NS, article studied factor but it was not significant; UNI, significant on univariant analysis only; OR, odds ratio (only given if results was significant).

The risk of LLNM from PTMC has not been well studied. One group found ETE (OR = 7.925), multifocality (OR = 3.560), and CLNM (OR = 2.334) to be predictors of LLNM (12). Another group found three ultrasonographic features (location in the upper pole [OR = 4.7], contact with the capsule [OR = 10.8], and calcifications [OR = 4.8)), as well as the presence of CLNM (OR = 6.9) to be predictive of LLNM (22). Multifocality was not found to be a predictor of LLNM, but ETE (OR = 16.066) was found to be a predictor of LLNM.

PTMC is considered to be an indolent disease due to its low recurrence rate and high overall survival, supported by the present findings of a low (3%) recurrence rate and no distant metastasis or death due to PTMC. However, LNM is still a risk, as we found that 14% of all patients (38/273) and 23% of patients with lymph nodes removed (38/163) had either CLNM or LLNM. Clinicians must be aware of this when planning treatment for PTMC, even if the PTMC was discovered incidentally, because the present data show that the risk of LNM in incidental PTMCs compared to known/suspected PTC is not significantly different (Table 2).

An additional concept that merits discussion is the extent of ETE and its potential impact on metastasis and recurrence. ETE was determined to be a significant predictor of both CLNM and LLNM. Of patients with adequate (>3 months) follow-up, 11 had ETE. Follow-up ranged from 6 to 88 months (median 40 months). Of these 11 patients, three (27%) recurred. Two of these had skeletal muscle invasion, while one had invasion into perithyroidal fibroadipose tissue only. Of the eight (73%) patients that did not recur, none had skeletal muscle invasion; all had invasion into perithyroidal fibroadipose tissue only (Fig. 2). Recent studies have suggested that so-called minimal ETE (defined as invasion into fibroadipose tissue or sternothyroid muscle) correlates with the presence of LNM but does not correlate with recurrence in multivariate analysis (23,24). As the majority of patients with ETE and adequate follow-up in this study did not recur, it is hypothesized that minimal ETE may not be a significant predictor of recurrence. The two patients with skeletal muscle invasion had a recurrence, whereas only 1/9 with invasion into fibroadipose tissue developed a recurrence. Perhaps peripherally located carcinomas gain access to lymphatic vessels more easily than centrally located carcinomas, and so-called minimal ETE may be an indicator of peripheral lymphatic access rather than a predictor of recurrence or death. Of the patients that had a recurrence, all had LNM at the time of thyroidectomy.

FIG. 2.

Low-power view of a peripherally located papillary thyroid microcarcinoma (A), demonstrating minimal extrathyroidal extension into adipose tissue (B); this carcinoma did not recur. Low-power view of a microcarcinoma (C) with extension into perithyroidal skeletal muscle (D); this carcinoma recurred (recurrence case #4). Low-power view of a microcarcinoma with equivocal extrathyroid extension into fibroadipose tissue (E); this carcinoma recurred (recurrence case #2). High-power view of an intrathyroidal carcinoma (F) that recurred (recurrence case #1). Color images available online at www.liebertpub.com/thy

It is acknowledged that there is still controversy regarding the long-term effectiveness of RAI and prophylactic CLND and whether these therapies increase overall survival or decrease recurrence in patients with PTMC. A recent publication found that prophylactic CLND does not increase postoperative complications such as hemorrhage, permanent hypoparathyroidism, or permanent recurrent laryngeal nerve injury, but a lack of prophylactic CLND does significantly increase the risk of recurrence (25). Additionally, evaluating the genetics of PTMC may further aid in prognostication and treatment decisions. Recent studies found that BRAF mutations in PTMC are associated with CLNM, occult contralateral carcinoma, or ETE (14,26,27). Certain morphologic features (infiltrative growth pattern, desmoplasia, cystic change, and “classic” PTC-like nuclei) may be associated with the presence of a BRAF mutation. However, on multivariate analysis, only the non-follicular variants were significantly associated with the presence of a BRAF mutation (27). Genetic evaluation and morphologic analysis of the current cohort are considered as next steps. However, these are beyond the scope and were not the main aims of this study.

Based on the present findings, a combination of clinical and pathologic information could be used to determine the appropriate treatment or follow-up plan for a patient. More aggressive treatment (such as RAI) or more frequent follow-ups could be considered if the patient is young (<45 years) or has a tumor with ETE or multifocality. All three of these factors are independent predictors of CLNM, which has been shown by some studies to decrease survival and increase recurrence in patients with PTMC (4,8). Conversely, older patients (≥45 years) with unifocal, intrathyroidal tumors are less likely to have LNM, and may be treated conservatively. Of 97 patients meeting these criteria (age ≥45 years with unifocal, intrathyroidal tumors), 48 had lymph node dissection and seven had LNM (6 LLNM and 1 CLNM). None of these patients had a recurrence. These findings suggest that there is no absolute combination of criteria that preclude LNM.

Some authors have suggested renaming a subset of PTMCs to “papillary thyroid microtumors” if they fulfill the so-called Porto criteria: patient age ≥19 years, unifocal, lack of extrathyroidal and intravascular invasion, lack of tall-cell features, and lack of metastasis (5). The authors support this concept, as long as the criteria remain strict. In the present cohort, no patient meeting the Porto criteria experienced a recurrence.

The overall recurrence rate in this study was 3%, of which most (5/6) occurred before 2.5 years of follow-up. Long-term follow-up of PTC patients in the authors' institution (not restricted to microcarcinomas) found that the mean time to recurrence was eight years, with a median of 27 years of follow-up (21). It is possible that PTMC may also recur after a decade, and that some of the patients in the cohort with shorter follow-up times will eventually have a recurrence. As only 15 of the patients had eight or more years of follow-up, these patients will need continued following in order to for there to be adequate power to predict recurrence.

Another limitation to this study is the relatively small number of patients with resected lymph nodes (163/273), which resulted in broad confidence intervals in the multivariate logistic regression analysis. Additionally, not all lymph node dissections were complete central compartment or lateral neck dissections; some consisted of the removal of solitary or random lymph nodes. The rate of LNM may be underrepresented due to the incomplete histologic evaluation of regional lymph nodes in all patients. Despite these limitations, these study results are consistent with the odds ratios and risk factors for LNM found in other studies.

In conclusion, PTMC can metastasize to central or lateral lymph nodes, which may increase the risk of recurrence and perhaps decrease survival. The presence or absence of unfavorable clinical and pathologic features described in this study (age <45 years, multifocality, ETE) may help direct treatment and follow-up decisions for PTMC patients.

Footnotes

Acknowledgments

Funding provided internally by the Department of Pathology and Pritzker School of Medicine, The University of Chicago, Chicago, IL.

Author Disclosure Statement

None.

References

Surveillance, Epidemiology, and End Results Program Turning Cancer Data into Discovery. Cancer of the thyroid. Available at: http://seer.cancer.gov/statfacts/html/thyro.html (accessed July 7, 2014 ).

MedlinePlus Medical Encyclopedia. U.S National Library of Medicine. Thyroid cancer—papillary carcinoma. Available at: www.nlm.nih.gov/medlineplus/ency/article/000331.htm (accessed July 7, 2014 ).

National Cancer Institute. Thyroid cancer treatment (PDQ^®). Available at: www.cancer.gov/cancertopics/pdq/treatment/thyroid/HealthProfessional/page4 (accessed July 7, 2014 ).

X-M

, Wan

, Sippel

, Chen

. 2011. Should all papillary thyroid microcarcinomas be aggressively treated? An analysis of 18,445 cases. Ann Surg, 254:653–660.

Rosai

, LiVolsi

, Sobrinho-Simoes

, Williams

. 2003. Renaming papillary microcarcinoma of the thyroid gland: the Porto proposal. Int J Surg Pathol, 11:249–251.

Chow

S-M

, Law

SCK

, Chan

JKC

, Au

S-K

, Yau

, Lau

W-H

. 2003. Papillary microcarcinoma of the thyroid: prognostic significance of lymph node metastasis and multifocality. Cancer, 98:31–40.

Lim

, Choi

, Yoon

Y-H

, Kim

E-H

, Koo

. 2009. Central lymph node metastases in unilateral papillary thyroid microcarcinoma. Br J Surg, 96:253–257.

Liu

, Wang

, Yi

, Wang

, Huang

. 2014. Risk factors for central lymph node metastasis of patients with papillary thyroid microcarcinoma: a meta-analysis. Int J Clin Exp Pathol, 7:932–937.

Sugitani

, Fujimoto

. 1999. Symptomatic versus asymptomatic papillary thyroid microcarcinoma: a retrospective analysis of surgical outcome and prognostic factors. Endocr J, 46:209–216.

10.

Pisanu

, Reccia

, Nardello

, Uccheddu

. 2009. Risk factors for nodal metastasis and recurrence among patients with papillary thyroid microcarcinoma: differences in clinical relevance between nonincidental and incidental tumors. World J Surg, 33:460–468.

11.

Hay

, Hutchinson

, Gonzalez-Losada

, McIver

, Reinalda

, Grant

, Thompson

, Sebo

, Goellner

. 2008. Papillary thyroid microcarcinoma: a study of 900 cases observed in a 60-year period. Surgery, 144:980–987; discussion 987–988.

12.

Zhang

, Wei

, Ji

, Zhu

, Wang

, Huang

, Shen

, Li

, Wu

. 2012. Risk factors for neck nodal metastasis in papillary thyroid microcarcinoma: a study of 1066 patients. J Clin Endocrinol Metab, 97:1250–1257.

13.

Kuo

S-F

, Lin

S-F

, Chao

T-C

, Hsueh

, Lin

K-J

, Lin

J-D

. 2013. Prognosis of multifocal papillary thyroid carcinoma. Int J Endocrinol, 2013:1–6.

14.

Yang

, Chen

, Jiang

, Chen

, Jin

, Guo

, Zhang

, Ye

. 2014. Prediction of central compartment lymph node metastasis in papillary thyroid microcarcinoma. Clin Endocrinol (Oxf), 81:282–288.

15.

Chéreau

, Buffet

, Trésallet

, Tissier

, Golmard

, Leenhardt

, Menegaux

. 2014. Does extracapsular extension impact the prognosis of papillary thyroid microcarcinoma?. Ann Surg Oncol, 21:1659–1664.

16.

Ahn

, Sohn

, Jeon

, Jeong

. 2014. Clinical impact of microscopic extrathyroidal extension in patients with papillary thyroid microcarcinoma treated with hemithyroidectomy. J Endocrinol Invest, 37:167–173.

17.

Roti

, Rossi

, Trasforini

, Bertelli

, Ambrosio

, Busutti

, Pearce

, Braverman

, Degli Uberti

. 2006. Clinical and histological characteristics of papillary thyroid microcarcinoma: results of a retrospective study in 243 patients. J Clin Endocrinol Metab, 91:2171–2178.

18.

American Joint Committee on Cancer 2010 Cancer Staging Manual. Seventh edition. Springer, New York.

19.

Zhou

, Gao

, Zhang

, Yang

, Guo

, Zhang

, Wang

. 2012. Factors predictive of papillary thyroid micro-carcinoma with bilateral involvement and central lymph node metastasis: a retrospective study. World J Surg Oncol, 10:67.

20.

, Jiang

, Han

. 2014. [Necessity of central lymph node dissection in management of papillary thyroid microcarcinoma]. Lin Chuang Er Bi Yan Hou Tou Jing Wai Ke Za Zhi, 28:362–365.

21.

Grogan

, Kaplan

, Cao

, Weiss

, Degroot

, Simon

, Embia

, Angelos

, Kaplan

, Schechter

. 2013. A study of recurrence and death from papillary thyroid cancer with 27 years of median follow-up. Surgery, 154:1436–1446; discussion 1446–1447.

22.

Kwak

, Kim

, Son

, Chung

, Park

, Nam

. 2009. Papillary microcarcinoma of the thyroid: predicting factors of lateral neck node metastasis. Ann Surg Oncol, 16:1348–1355.

23.

Woo

, Sung

, Choi

, Kim

, Shong

, Kim

, Hong

, Song

. 2015. Clinicopathological significance of minimal extrathyroid extension in solitary papillary thyroid carcinomas. Ann Surg Oncol, 22:S728–733.

24.

Kim

, Lee

, Choi

, Lim

, Lee

, Cho

, Lee

, Yi

. 2010. The clinical importance of minimal extrathyroid extension on tumor recurrence in patients with papillary thyroid carcinoma. Endocrinol Metab, 25:340–346.

25.

Zhang

, Liu

, Gao

, Zheng

. 2015. The clinical prognosis of patients with cN0 papillary thyroid microcarcinoma by central neck dissection. World J Surg Oncol, 13:138.

26.

Zhou

, Zhang

, Gao

, Dai

, Yang

, Zhang

, Wang

. 2012. Preoperative BRAF mutation is predictive of occult contralateral carcinoma in patients with unilateral papillary thyroid microcarcinoma. Asian Pac J Cancer Prev, 13:1267–1272.

27.

Virk

, Van Dyke

, Finkelstein

, Prasad

, Gibson

, Hui

, Theoharis

, Carling

, Roman

, Sosa

, Udelsman

, Prasad

. 2013. BRAFV600E mutation in papillary thyroid microcarcinoma: a genotype-phenotype correlation. Mod Pathol, 26:62–70.

28.

Chang

, Kim

, Lee

, Bae

, Son

. 2015. Should central lymph node dissection be considered for all papillary thyroid microcarcinoma?. Asian J Surg, 2015 Apr 22. DOI: 10.1016/j.asjsur.2015.02.006. [Epub ahead of print].

29.

, Lv

, Wang

, Dai

. 2014. Risk factors for predicting central lymph node metastasis in papillary thyroid microcarcinoma. Int J Clin Exp Pathol, 7:6199–6205.

30.

Pisanu

, Saba

, Podda

, Reccia

, Uccheddu

. 2015. Nodal metastasis and recurrence in papillary thyroid microcarcinoma. Endocrine, 48:575–581.

31.

Park

, Roh

, Lee

, Baek

, Gong

, Cho

, Choi

, Nam

, Kim

. 2014. Risk factors for central neck lymph node metastasis of clinically noninvasive, node-negative papillary thyroid microcarcinoma. Am J Surg, 208:412–418.

32.

Zhang

, Liu

, Gao

, Zheng

. 2015. Risk factors for nodal metastasis in cN0 papillary thyroid microcarcinoma. Asian Pac J Cancer Prev, 16:3361–3363.