High Grade Differentiated Follicular-cell Derived Thyroid Carcinoma Versus Poorly Differentiated Thyroid Carcinoma: a Clinicopathologic Analysis of 41 Cases

Abstract

Abstract Background Criteria overlap for separating between malignant follicular epithelial cell derived thyroid gland neoplasms with high grade features of increased mitoses and tumor necrosis but lacking anaplastic histology. Patterns of growth (insular), nuclear features, tumor necrosis, and various cutoff of mitotic activity are suggested, but a reproducible Ki-67-based labelling index has not been established. Materials All cases (n=41) diagnosed as poorly differentiated thyroid carcinoma (PDTC) or high grade differentiated follicular cell-derived thyroid carcinoma (HGDFCDTC) were reviewed, with histologic features, mitotic figure counts, and Ki-67 labelling index reviewed on cases within Southern California Permanente Medical Group from 2010 to 2021 to detect any outcome differences. Results There were 17 HGDFCDTC (nine papillary thyroid carcinoma; eight oncocytic follicular thyroid carcinoma), median age 64 years, affecting nine females and eight males. Tumors were large (median, 6.0 cm), usually unifocal (n=13), with only one tumor lacking invasion. Tumor necrosis was present in all; median mitotic count was 5/2 mm2 (standard deviation [SD] 3.4), with a median Ki-67 labelling index of 8.3% (median 684 mitoses/8,708 tumor nuclei/2 mm2). Three patients had metastatic disease at presentation, with additional metastases in four patients (41.2% developed metastases); 11 were alive (n=10) or dead (n=1) without evidence of disease (median 21.2 months); with the remaining six patients alive (n=4) or dead (n=2) with metastatic disease (median 25.8 months). Criteria associated with an increased risk of developing metastatic disease: widely invasive tumors; age ≥55 years; male; advanced tumor size and stage; extrathyroidal extension; but not increased mitotic rate or higher labelling index. There were 24 PDTC, median age 57.5 years, affecting 13 females and 11 males. Tumors were large (median, 6.9 cm), with 50% part of multifocal disease, but three tumors lacking invasion. Insular/trabecular/solid architecture was seen in all tumors; tumor necrosis was present in 23; median mitotic count was 6/2 mm2 (SD 10.6), with a median Ki-67 labelling index of 6.9% (median 620 mitoses/10,278 tumor nuclei/2 mm2). Five patients had metastatic disease at presentation, with additional metastases in 3 patients (29.2% developed metastases); 16 were alive (n=15) or dead (n=1) without evidence of disease (median, 48.1 months); with the remaining 8 patients alive (n=3) or dead (n=5) with metastatic disease (median, 22.4 months). Criteria associated with an increased risk of developing metastatic disease: widely invasive tumors; male; advanced tumor size and stage; extrathyroidal extension; but not increased mitotic rate or higher labelling index. Conclusion: HGDFCDTC show tumor necrosis, a median Ki-67 labelling index of 8.3%, with a high percentage (41%) of patients developing metastatic disease. PDTC presents slightly with large tumors, often in a background of multifocal tumors, with tumor necrosis nearly always seen, with a similar median Ki-67 labelling index of 6.9, with 29% of patients developing metastatic disease. Separation between groups is meaningful as early metastatic disease is relatively common, but mitotic counts/labelling indices are not different between the groups nor able to potentially risk stratify development of metastatic disease. It is hereby proposed to use ≥3 mitoses/2 mm2 as the criterion for both tumor types, with/without tumor necrosis for the HGDFCDTC group and keep the PDTC criteria unchanged.