Summary Among thyroid papillary carcinomas (PTCs), the follicular variant is the most common and includes encapsulated forms (EFVPTCs). Noninvasive EFVPTCs have very low risk of recurrence or other ...adverse events and have been recently proposed to be designated as noninvasive follicular thyroid neoplasm with papillary-like nuclear features or NIFTP , thus eliminating the term carcinoma . This proposal is expected to significantly impact the risk of malignancy associated with the currently used diagnostic categories of thyroid cytology. In this study, we analyzed the fine needle aspiration biopsy (FNAB) cytology features of 96 histologically proven NIFTPs and determined how the main nuclear features of NIFTP correlate between cytological and histological samples. Blind review of FNAB cytology from NIFTP nodules yielded the diagnosis of “follicular neoplasm” (Bethesda category IV) in 56% of cases, “suspicious for malignancy” (category V) in 27%, “atypia of undetermined significance/follicular lesion of undetermined significance” (category III) in 15%, and “malignant” (category VI) in 2%. We found good correlation (κ = 0.62) of nuclear features between histological and cytological specimens. NIFTP nuclear features (size, irregularities of contours, and chromatin clearing) were significantly different from those of benign nodules but not from those of invasive EFVPTC. Our data indicate that most of the NIFTP nodules yield an indeterminate cytological diagnosis in FNAB cytology and nuclear features found in cytology samples are reproducibly identified in corresponding histology samples. Because of the overlapping nuclear features with invasive EFVPTC, NIFTP cannot be reliably diagnosed preoperatively but should be listed in differential diagnosis of all indeterminate categories of thyroid cytology.
Summary Gastric cancer shows intratumoral heterogeneity for human epidermal growth factor receptor 2 expression. We evaluated whether the number of tissue blocks analyzed or the antibodies used may ...influence the immunohistochemical results in gastrectomy specimens. Clinicopathologic data from 148 patients receiving gastric surgery for cancer were collected. One tissue block for each of 88 primary tumors and 60 paired primary tumors and metastases was examined for human epidermal growth factor receptor 2 status by immunohistochemistry using 3 different antibodies (HercepTest, CB11, and 4B5) and by fluorescent in situ hybridization. Two additional tissue blocks of the primary tumor were tested by immunohistochemistry if the results were negative on the first tissue block. The concordance among the 3 antibodies was 94.5% (testing 1 tissue block). Two cases showed a clinically significant discrepancy between primary tumor (score 0) and lymph nodes metastases (score 3+). Additional block analysis increased both the sensitivity (from 63% to 83%) and the accuracy (from 91% to 94%) of immunohistochemistry as compared with fluorescent in situ hybridization. The multiblock approach could potentially identify a greater number of human epidermal growth factor receptor 2–positive gastric cancers, particularly those with higher levels of intratumor heterogeneity. In turn, human epidermal growth factor receptor 2 positivity correlated with a worse prognosis ( P = .011) and was an independent variable in multivariate analysis (hazard ratio, 1.57). In conclusion, testing more than 1 tissue block of cancer from specimens of gastric resection provides a more reliable human epidermal growth factor receptor 2 assessment regardless of the antibody used.