Mitotic count of fine needle aspiration material of gastrointestinal stromal tumours of the stomach underestimates actual mitotic count.

INTRODUCTION: A mitotic count is required for histological grading in resections of gastrointestinal stromal tumours (GISTs). However, no consensus on the utility of mitotic count in fine needle aspiration (FNA) GIST material currently exists. This study examines the relationship between mitotic counts of FNAs and subsequent resections of GISTs of the stomach. MATERIALS AND METHODS: We identified 39 cases of GISTs of the stomach diagnosed via FNA at our institution between January 1, 2014, and December 31, 2019, with subsequent resection. We noted if rapid on-site evaluation (ROSE) was performed. Cell block (CB) material from FNAs was analysed for total area, percentage containing neoplastic cells, and number of mitoses. We compared the mitotic counts in CBs and subsequent resections with t tests. RESULTS: ROSE was performed in 82% of cases and called adequate every time. Mean values for total CB area, neoplastic material percentage, and area of neoplastic cells were 54.7 mm2 (range 1-986), 45% (range 10%-90%), and 19.2 mm2 , respectively; 27 cases (69%) had greater than 50 high powered fields of GIST material in the CB. Mean numbers of mitoses per 5 mm2 were 0.38 (range 0-11) for CBs versus 5.92 (range 0-70) for resections (P < 0.05). CONCLUSION: At our institution, ROSE adequacy of spindle cell lesions focuses on diagnosing GIST rather than obtaining adequate material for histological grading. Mitotic figures were statistically lower in FNA CB material than subsequent resections, and using mitotic counts from CB material may underestimate the histological grade of GISTs of the stomach.

Perilipin 5 regulates islet lipid metabolism and insulin secretion in a cAMP-dependent manner: implication of its role in the postprandial insulin secretion.

Elevation of circulating fatty acids (FA) during fasting supports postprandial (PP) insulin secretion that is critical for glucose homeostasis and is impaired in diabetes. We tested our hypothesis that lipid droplet (LD) protein perilipin 5 (PLIN5) in ß-cells aids PP insulin secretion by regulating intracellular lipid metabolism. We demonstrated that PLIN5 serves as an LD protein in human islets. In vivo, Plin5 and triglycerides were increased by fasting in mouse islets. MIN6 cells expressing PLIN5 (adenovirus [Ad]-PLIN5) and those expressing perilipin 2 (PLIN2) (Ad-PLIN2) had higher [(3)H]FA incorporation into triglycerides than Ad-GFP control, which support their roles as LD proteins. However, Ad-PLIN5 cells had higher lipolysis than Ad-PLIN2 cells, which increased further by 8-Br-cAMP, indicating that PLIN5 facilitates FA mobilization upon cAMP stimulation as seen postprandially. Ad-PLIN5 in islets enhanced the augmentation of glucose-stimulated insulin secretion by FA and 8-Br-cAMP in G-protein-coupled receptor 40 (GPR40)- and cAMP-activated protein kinase-dependent manners, respectively. When PLIN5 was increased in mouse ß-cells in vivo, glucose tolerance after an acute exenatide challenge was improved. Therefore, the elevation of islet PLIN5 during fasting allows partitioning of FA into LD that is released upon refeeding to support PP insulin secretion in cAMP- and GPR40-dependent manners.