The prognostic value of primary tumor size in papillary and follicular thyroid carcinoma

A comparative analysis


  • Andreas Machens M.D.,

    Corresponding author
    1. Department of General, Visceral, and Vascular Surgery, Martin-Luther-University Halle-Wittenberg, Halle/Saale, Germany
    • Department of General, Visceral and Vascular Surgery, Martin-Luther-University Halle-Wittenberg, Ernst-Grube-Straße 40, D-06097 Halle/Saale, Germany
    Search for more papers by this author
    • Fax: (011) 49-345-557-2551

  • Hans-Jürgen Holzhausen M.D.,

    1. Department of Pathology, Martin-Luther-University Halle-Wittenberg, Halle/Saale, Germany
    Search for more papers by this author
  • Henning Dralle M.D.

    1. Department of General, Visceral, and Vascular Surgery, Martin-Luther-University Halle-Wittenberg, Halle/Saale, Germany
    Search for more papers by this author



A delay in the diagnosis of differentiated thyroid carcinoma often leads to larger tumors, higher prevalence rates of distant metastasis, and earlier cause-specific deaths. Threshold tumor diameters for extrathyroidal growth, lymph node spread, and distant metastasis in papillary (PTC) and follicular thyroid carcinoma (FTC) remain to be defined.


A comparative correlation of primary tumor size and extrathyroidal growth, lymph node spread, and distant metastasis was performed for 500 institutional patients who received surgery for PTC or FTC.


There were 366 patients with PTC (73.2%) and 134 patients with FTC (26.8%). Multifocality (23.5% vs. 9.0%; P < 0.001) and lymph node metastasis (40.2% vs. 19.4%; P < 0.001) were more common in the patients with PTC than in those with FTC. Patients with FTC were older at first diagnosis (51.6 vs. 47.0 years; P = 0.01) compared with the patients with PTC. The FTC tumors were almost twice as large (39.9 vs. 20.6 mm; P < 0.001), and patients had a higher prevalence of distant metastasis (17.9% vs. 6.3%; P < 0.001). When primary tumor diameter was accounted for, cumulative risks of extrathyroidal growth and lymph node metastasis were higher in patients with PTC than in patients with FTC (P < 0.001; log-rank test). In striking contrast, the cumulative risk of distant metastasis was the same for PTC and FTC tumors of equal size (P = 0.89; log-rank test) and increased once the primary tumor size was > 20 mm. Pulmonary metastasis was an earlier event than bone metastasis.


The data suggested that earlier intervention is warranted to keep suspicious thyroid nodules from growing > 20 mm (or greater than T1) and spreading to distant organs. Cancer 2005. © 2005 American Cancer Society.

Sharing derivation from a common progenitor cell, papillary (PTC) and follicular thyroid carcinoma (FTC), the two most prevalent forms of thyroid carcinoma, are collectively referred to as differentiated thyroid carcinoma. The single most important determinant of survival in this condition is distant metastasis.1, 2 In the presence of distant metastasis, the median survival is estimated to be 4.1 years, and the 10-year disease-specific survival rate is 26%.3 Patients with metastasis to bone fare even worse.4 Mortality rates are similar for PTC and FTC when patients with distant metastasis at diagnosis are excluded.5

In clinical practice, however, the presentation of PTC and FTC differs markedly. For example, patients with follicular carcinomas are on average 3–6 years older than those who have papillary carcinomas (38.7 vs. 34.9 yrs,5 47 vs. 44 yrs,6 55 vs. 50 yrs,7 and 60 vs. 54 yrs8), and their tumors are 6–18 mm larger (29 vs. 23 mm,5 30 vs. 17 mm,6 35 vs. 20 mm,9). Fine-needle aspiration cytology often does not enable a clear discrimination of follicular carcinoma from follicular adenoma.10 A delay in tumor diagnosis alone could explain the disparity between the tumor entities in the prevalence of distant metastasis and, ultimately, survival.

In patients with differentiated thyroid carcinoma, primary tumor size may determine an individual's risk for developing extrathyroidal growth, lymph node spread, and distant metastasis. To delineate threshold tumor diameters for all 3 conditions, the current comparative study of 500 patients with PTC and FTC was conducted.


Selection of the Study Population

Between November 1994 and March 2004, 426 consecutive patients with PTC and 185 consecutive patients with FTC received surgery at the Martin-Luther-University Halle-Wittenberg (Halle/Saale, Germany). Information on primary tumor size was unavailable for 111 patients, reducing the study population to 500 of 611 patients (81.8%). Of these remaining patients, 366 patients had PTC and 134 patients had FTC. As previously demonstrated in a subset of 417 of all 611 institutional patients, unavailablility of tumor diameter in differentiated thyroid carcinoma is associated with extrathyroidal growth and tracheal invasion, presumably due to piecemeal removal of the tumor at primary thyroid surgery.11 To accommodate this limitation, cumulative risks in the current study were related to tumor size.

Extent of Surgery

Of the 500 patients, 150 patients (30%) underwent primary surgery and 350 patients (70%) underwent reoperations. Two-thirds (235 of 350) of these reoperations accounted for completion thyroidectomies with central lymph node dissections whereas the remainder (115 of 350) were interventions for recurrent tumors. Therefore, 461 patients (92.2%) received a total thyroidectomy. A total of 414 patients (82.8%) underwent systematic dissections of the central lymph node compartment, which extends vertically from the hyoid bone to the thoracic inlet and horizontally between the carotid sheaths.12 The lateral cervical compartments were dissected in 135 (27.0%; ipsilateral) and 45 patients (9.0%; contralateral), and the mediastinal compartment in 26 patients (5.2%). All clinical interventions represented standard practice of care and agreed with the practice guidelines of the German Society of Surgery.13 Before surgery, informed consent was obtained for each of the above procedures. All specimens were subjected to pathologic examination.

Pathologic Examination and Tumor Staging

After macroscopic evaluation by the pathologist, the entire thyroid gland was divided vertically to separate the left and right lobes. The thyroid halves were then sectioned horizontally from the superior to the inferior pole, as described previously.14 After fixation in formalin, the whole thyroid gland was embedded in paraffin. Soft tissue and lymph nodes were processed separately. Conventional staining (hematoxylin and eosin) and, where appropriate, thyroglobulin immunohistochemistry were performed on every surgical specimen, using the standard avidin-biotin complex peroxidase approach. The diagnosis of PTC and FTC was based on the World Health Organization histologic classification of tumors.15 Differentiated tumors with one or more anaplastic foci were regarded as undifferentiated thyroid carcinomas and were not included in the current study. Primary tumor size was ascertained at primary surgery by direct measurements on the resected thyroid specimen. When multiple tumors were present, only the size of the largest thyroid tumor was taken. Although a diagnosis of lymph node metastasis always required pathologic confirmation, this need was waived for distant metastasis when there was such unequivocal evidence on ultrasonography (liver), computed tomography scans, magnetic resonance imaging scans, positron emission tomography scans, scintiscan, or any combination of these imaging modalities.

Statistical Analysis

Categorical and continuous data were tested on univariate analysis using the two-tailed Fisher exact test and the two-tailed exact Mann–Whitney–Wilcoxon rank sum test, respectively. Multiple testing was adjusted for using the Bonferroni method. For comparative analysis of cumulative risks between both tumor entities, the Kaplan–Meier method (log-rank test)16 was used. The level of significance was set at P = 0.05.


Patient Demographics and Pathologic Features According to Tumor Entity

Table 1 specifies the study population stratified by type of thyroid malignancy: 366 patients (73.2%) had PTC and 134 patients (26.8%) had FTC. PTCs tumors were more often multifocal (23.5% vs. 9.0%), and involved more frequently lymph nodes (40.2% vs. 19.4%) compared with FTC tumors. Patients with follicular carcinomas were almost 5 years older at receipt of primary surgery than those with papillary carcinomas (51.6 vs. 47.0 yrs). Follicular tumors were almost twice as large as papillary tumors (39.9 vs. 20.6 mm), and were more frequently associated with distant metastasis to bone (7.5% vs. 1.4%), lung (12.7% vs. 5.7%), and distant metastasis overall (17.9% vs. 6.3%) at the time of the most recent surgery. Reoperations (completion thyroidectomies with central lymph node dissections and interventions for recurrent tumors combined) involved more commonly FTC than PTC (78.4% vs. 66.9%). No significant differences between the tumors were noted on univariate analysis with regard to patient gender (female gender: 72.7% vs. 70.1%) and extrathyroidal growth (25.4% vs. 17.2%). After multiple testing was corrected, only multifocality, primary tumor diameter, lymph node metastasis, distant metastasis overall, and bone metastasis were retained as significant discriminatory factors between the two tumors.

Table 1. Clinicopathologic Features By Tumor Entity
FactorsPTC (%) (n = 366)FTC (%) (n = 134)P valuea
  • PTC: papillary thyroid carcinoma; FTC: follicular thyroid carcinoma.

  • a

    Two-tailed exact Mann–Whitney–Wilcoxon rank-sum test and Fisher exact test, respectively.

  • b

    95% confidence intervals are shown in brackets.

  • c

    Significant after Bonferroni correction for multiple testing.

  • d

    As ascertained at the most recent surgery.

Mean patient age (yrs) at first diagnosis47.0 [45.2; 48.8]b51.6 [48.7; 54.5]b0.01
Mean tumor diameter (mm)20.6 [18.8; 22.4]b39.9 [35.8; 44.1]b<0.001c
Multifocal growth23.59.0<0.001c
Extrathyroidal growth25.417.20.06
Lymph node metastasis40.219.4<0.001c
Distant metastasisd   

Cumulative Risks of Extrathyroidal Growth, Lymph Node Metastasis, and Distant Metastasis According to Tumor Diameter

Cumulative frequency plots provide an estimate of the probability, or risk, that an individual will have developed the condition under investigation by a certain tumor diameter.17 Unlike average rates, cumulative rates are more representative of an individual's risk of having the condition of interest. Figures 1 and 2 depict the cumulative risks of extrathyroidal growth, lymph node metastasis, and distant metastasis according to primary tumor diameter. In all three categories, the cumulative risks increased linearly with increasing tumor diameter.

Figure 1.

Cumulative risks of extrathyroidal growth and lymph node metastasis. PTC: papillary thyroid carcinoma; FTC: follicular thyroid carcinoma.

Figure 2.

Cumulative risks of distant metastasis overall pulmonary metastasis and bone metastasis. PTC: papillary thyroid carcinoma; FTC: follicular thyroid carcinoma.

For cumulative risks of extrathyroidal growth and lymph node metastasis (Fig. 1), the slope of the curves for PTC was steeper than the slope of the curves for FTC (P < 0.001; log-rank test), indicating earlier extrathyroidal extension and lymph node metastasis for patients with PTC. Another difference concerned the position of the anchoring point on the ordinate (x-intercept) of the curves: for PTC at a threshold tumor diameter of 5 mm, and for FTC at a threshold tumor diameter of 20 mm.

For the cumulative risk of distant metastasis overall (Fig. 2), the curves for both tumors were virtually congruent when primary tumor diameter was accounted for (P = 0.89; log-rank test). There was no difference between papillary and follicular carcinomas when the cumulative risks of distant metastasis to the lung (P = 0.35; log-rank test) and bone (P = 0.36; log-rank test) were analyzed separately (Fig. 2). Regardless of tumor entity, the slope of the curves for pulmonary metastasis was steeper than the one for bone metastasis. The anchoring point on the ordinate (x-intercept) of the curve for pulmonary metastasis was located left (at a threshold tumor diameter of 20 mm) of the anchoring point of the curve for bone metastasis (at a threshold tumor diameter of 30–40 mm), suggesting that pulmonary metastasis is an earlier event than bone metastasis in both thyroid tumors.


To our knowledge, this is the first demonstration of similar rates of disease progression from equally large PTC and FTC to distant metastasis. In the current series, primary thyroid tumors did not progress to distant metastasis until they were > 20 mm in greatest diameter. This critical threshold diameter of 20 mm may be useful in stratifying thyroid nodules according to the risk of distant metastasis in differentiated thyroid carcinoma. For example, although thyroid nodules ≤ 20 mm are rather low risk and may warrant expectant observation, thyroid nodules > 20 mm may require a more thorough and complete workup, where applicable with close follow-up intervals. It is noteworthy in this regard that the current 2002 TNM classification18 for thyroid carcinoma increased the line of separation between the T1 and T2 classifications from 10 mm (1997 TNM classification19) to 20 mm.

Stratification of Thyroid Nodules by Primary Tumor Diameter

Considering the high prevalence of palpable thyroid nodules of 4–7% in the adult U.S. population,20 cost-effective, risk-reduction strategies are needed. Our data suggest that routine stratification of thyroid nodules by primary tumor diameter may pave the way for timely surgical intervention, preventing systemic disease in an additional subset of patients with differentiated thyroid carcinomas. Thyroid nodules > 20 mm make up one-third of thyroid nodules.21 Primary tumor diameter, as well as lymph node enlargement, is easily ascertained by use of high-resolution ultrasonography22 which can also serve to guide fine-needle aspiration. Definitive exclusion of malignancy by fine-needle aspiration cytology should be sought in the one-third of patients whose thyroid nodules are > 20 mm. When malignancy cannot be definitively excluded, especially in follicular neoplasia on fine-needle aspiration cytology and nonfunctioning nodules on scanning, the patient should be offered surgery. Overreliance on fine-needle aspiration cytology findings in patients with thyroid carcinoma, sometimes even despite clinical evidence suggesting malignancy, can defer surgical treatment by 28 months and adversely affect outcome.23

Meticulous workup of suspect thyroid nodules may advance the diagnosis of differentiated thyroid carcinoma, allowing for timely surgical intervention before distant metastasis ensues. This principle is epitomized by the delay of 3–6 years in diagnosing FTC relative to PTC. This delay in diagnosis may allow follicular carcinomas to progress farther than their papillary counterparts, accounting for the higher prevalence of large follicular tumors with distant metastasis and hence poorer survival rates. A heightened index of suspicion is the key element of success in diagnosing thyroid carcinomas early. Stratification of thyroid nodules by tumor diameter may contribute to a more evidence-based practice of screening for thyroid malignancy.