Dual-color immunocytochemistry (Ki-67 with LCA) for precise grading of pancreatic neuroendocrine tumors with applicability to small biopsies and cell blocks.

Ki-67 (MIB-1) immunostaining to quantify the proliferative index of neuroendocrine tumors (NETs) has been recommended (especially for small biopsies). However, this has a number of challenges with nonrepresentative Ki-67 index due to interference by Ki-67 immunoreactive proliferating lymphocytes infiltrating the tumor and also some proliferating stromal cells including endothelial cells in the background. Our pilot project showed that dual-color immunostaining with inclusion of leukocyte common antigen (LCA) (Ki-67: nuclear brown; LCA: cytoplasmic red) can facilitate the weeding out of lymphocyte interference. We analyzed the results with 23 surgical cases of pancreatic NETs. This was followed by poststudy examination of 11 cases of endoscopic ultrasound-guided fine-needle aspiration of the pancreatic NETs (PanNETs) to evaluate the findings of the study. Dual-color immunostaining for Ki-67 with LCA increased the precision of quantifying Ki-67 index, due to ability to exclude LCA immunoreactive lymphocytes. Other nontumor Ki-67 immunoreactive cells such as endothelial and stromal cells could be distinguished morphologically. Digital methods were also attempted, but this approach could not distinguish infiltrating lymphocytes and other cells in sections resulting in erroneous results. This study demonstrated that grading of PanNET can be performed with increased precision with dual-color Ki-67 immunostaining protocol standardized in this study. As evaluated on a few cytopathology cases, this protocol is especially useful for the evaluation of small biopsies and cell block sections of fine-needle aspiration biopsy material where 50 high-power fields cannot be evaluated but have >500 tumor cell nuclei.

Predicting histological subtypes of follicular variant of papillary thyroid carcinoma based on cytomorphology. Can cytomorphology optimize use of molecular testing?

INTRODUCTION: Follicular variant papillary thyroid carcinoma (FVPTC) can be further subclassified into one of 3 subtypes: non-invasive encapsulated FVPTC, invasive encapsulated FVPTC, and infiltrative FVPTC. Longitudinal and molecular studies have demonstrated that, in terms of both molecular profiles and prognosis, encapsulated FVPTC is comparable to follicular adenoma, invasive FVPTC to follicular carcinoma, and infiltrative FVPTC to classic PTC. To improve triaging and prevent overtreatment of patients with FVPTC, we sought to determine cytologic features likely to occur within each subtype. METHODS: A laboratory database search from 2010-2015 was conducted to identify patients with biopsy-proven FVPTC and prior fine-needle aspiration. Surgical specimens were reviewed to determine the appropriate subcategorization. Accompanying cytology reports were reviewed for features common in classic PTC and follicular neoplasms. RESULTS: Encapsulated variants were more likely to be graded as Bethesda category 4 compared with invasive or infiltrative variants. In contrast, infiltrative variants were more likely to be graded as Bethesda categories 5 and 6 compared with invasive or encapsulated variants. Compared with the encapsulated variant, infiltrative FVPTC was more likely to have nuclear pseudo-inclusions (31.82% versus 8.11%, P = 0.0468) and less likely to have microfollicular architecture (22.73% versus 54.05%, P = 0.0374). CONCLUSION: This study identified cytomorphologic differences between encapsulated and infiltrative FVPTC. With a higher threshold of suspicion for FVPTC, improved awareness of the differences between these subtypes and incorporation of molecular testing, it is likely that the Bethesda category can be revised and patient triaging can be significantly improved.