Large sporadic thyroid medullary carcinomas: predictive factors for lymph node involvement

2018 
Lymph node involvement (LNI) is one of the most important prognostic factors for poor survival in medullary thyroid carcinoma (MTC). At diagnosis, LNI is found in over 50% of sporadic MTCs, and especially in large tumours. Cervical lymph node dissection is therefore mandatory during MTC surgery. However, some large tumours (responsible for high preoperative basal calcitonin levels) are found to lack LNI, and can be cured definitely. Preoperative detection of these particular tumours might spare patients from undergoing extensive cervical dissection. The objective of the present retrospective study of a series of large sporadic MTCs was to identify clinical, biological and pathological factors that were predictive of LNI. Consecutive cases of large, sporadic MTCs (measuring at least 1 cm in diameter) were retrieved and reviewed. The levels of several mature microRNAs (miRs) in paraffin-embedded samples were assessed using qPCR. Of the 54 MTCs, 26 had LNI and 28 were pN0. Relative to pN0 patients, patients with LNI had a significant higher preoperative basal calcitonin level (p = 0.0074) and a greater prevalence of infiltrative margins (p < 0.0001), lymphovascular invasion (p = 0.0004), extrathyroidal extension (p < 0.0001), a higher pT stage (p = 0.0003) and more abundant desmoplastic stroma (p = 0.0006). Tumour expression levels of miR-21 (p = 0.0008) and miR-183 (p = 0.0096) were higher in the LNI group. The abundance of desmoplastic stroma (p = 0.007) and the miR-21 expression level (p = 0.0026) were independent prognostic factors for LNI. The abundance of desmoplastic stroma and high levels of miR-21 expression were strong indicators of LNI, and may thus help the surgeon to choose the extent of cervical lymph node dissection for large, sporadic MTCs with no preoperatively obvious LNI.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    31
    References
    10
    Citations
    NaN
    KQI
    []