Peripheral location and infiltrative margin predict invasive features of papillary thyroid microcarcinoma

in European Journal of Endocrinology

Correspondence should be addressed to Y S Jo; Email:

*(W K Lee and J Lee contributed equally to this work)

Restricted access


Tumor location in papillary thyroid microcarcinoma (PTMC) might determine tumor outgrowth from the thyroid gland. However, the clinical implications of tumor location and minimal extrathyroid extension (mETE) have not been well elucidated. We aimed to investigate the relationship between tumor location and mETE to predict the aggressiveness of PTMC.


A total of 858 patients with PTMC were grouped according to tumor location on ultrasonography: central (cPTMC) and peripheral PTMC (pPTMC). PTMC without mETE (PTMC-mETE(−)) was divided further according to margin shape: encapsulated (E−) or infiltrative (I−). To understand the molecular biologic characteristics of PTMC presenting with an I-margin and mETE, transcriptome data from TCGA-THCA were analyzed using Gene Set Enrichment Analysis (GSEA).


pPTMC (n = 807, 94.1%) accounted for the majority of cases; mETE was identified only in pPTMC (403/807; 49.9%). pPTMC-mETE(+) showed aggressive clinical characteristics that increased the odds ratio (OR) for lymph node metastasis (LNM). Interestingly, subgroup analysis of PTMC-mETE(−) revealed that the I-margin also increased the OR for LNM, independent of other clinical factors. GSEA of TCGA-THCA data suggested coordinated upregulation of genes related to epithelial-mesenchymal transition (EMT) in PTC with mETE. Immunohistochemical staining for laminin subunit gamma 2 (LAMC2), CD59, E-cadherin and vimentin showed that these markers of EMT were associated with progressive changes in E-margin PTMC-mETE(−), I-margin PTMC-mETE(−) and pPTMC-mETE(+).


mETE related to peripheral location of PTMC is an important predictor of tumor invasiveness, as is the I-margin, which presents with EMT features similar to mETE. I-margin PTMC-mETE(−) and pPTMC-mETE(+) might reflect the pattern of invasive PTMC.

Supplementary Materials

    • Supplementary Table 1. Thyroid volume (mL) of patients in this study compared with that of patients in previous studies
    • Supplementary Table 2. Multivariable analysis to determine relationships between tumor location and clinico-pathological factors and LNM in PTMC.
    • Supplementary Table 3. Multivariable analysis to determine the association between minimal ETE and LNM in TCGA-THCA
    • Supplementary Table 4. Multivariable logistic regression model (adjusted for all clinicopathological parameters) to determine the association between minimal ETE and tumor recurrence in TCGA-THCA.
    • Supplementary Figure 1. Classification of ultrasonographic tumor location. (A and B) Representative ultrasonographic axial images of cPTMC (A) and pPTMC (B). Abbreviations: cPTMC, central papillary thyroid microcarcinoma; pPTMC, peripheral papillary thyroid microcarcinoma.
    • Supplementary Figure 2. Clinicopathological implications of mETE in TCGA-THCA. (A and B) Comparison of tumor size (A) and LNM (B) according to ETE status (two-tailed Student’s t test, one-way ANOVA, or Chi-squared test). (C) Kaplan-Meier survival analysis according to ETE status. Data in A represent the mean ± SD. Missing data were excluded from each analysis. ns = non-significant, * = P < 0.05, **** = P < 0.0001. Abbreviations: EMT, epithelial-mesenchymal transition; ETE, extrathyroidal extension; LNM, lymph node metastasis; DM, distant metastasis; THCA, Thyroid Cancer; mETE, minimal ETE; gETE, gross ETE.
    • Supplementary Figure 3. Clinico-pathological implications of mETE in PTMC from TCGA-THCA. (A) Gene set related to EMT was coordinately up-regulated in PTC with mETE. (B) Frequent LNM in PTC with mETE. (C–F) Comparison of tumor size (C), tumor recurrence (D), age (E), and gender ratio (F) according to the presence of mETE. Continuous data were compared using a two-tailed Student’s t test, and categorical data were compared using the Chi-squared or Fisher’s exact test. Data represent the mean ± SD. ns = non-significant, * = P < 0.05, ** = P < 0.01.
    • Supplementary Figure 4. Representative markers of EMT in TCGA-THCA. (A) Heatmaps indicating the candidate genes, LAMC2 and CD59, in PTC and PTMC TCGA-THCA. Asterisks indicate genes showing significant differences in expression according to the presence of minimal ETE. (B) Expression of mRNA encoding LAMC2 and CD59 in normal and PTC tissue. Error bars indicate the mean ± SD. Continuous variables were analyzed using a two-tailed Student’s t test. ns = non-significant, **** = P < 0.0001.


     European Society of Endocrinology

Sept 2018 onwards Past Year Past 30 Days
Abstract Views 1270 1270 138
Full Text Views 109 109 6
PDF Downloads 66 66 2