Spindle cell neoplasms of the upper gastrointestinal tract, hepatobiliary tract, and pancreas by fine needle aspiration: A single institutional experience of 15 years with follow-up data.

BACKGROUND: The diagnosis of spindle cell neoplasms (SCN) of the upper gastrointestinal (GI) tract, hepatobiliary tract, and pancreas detected by fine needle aspiration (FNA) is challenging. We describe a single-center experience of these samples with follow-up data and characterization of the morphologic findings. METHODS: We retrospectively reviewed pathology records for all FNAs diagnostic for or suggestive of SCN on esophagus, stomach, small bowel, liver, and pancreas in a 15 year period. All cases with at least 6 month follow-up were included. Surgical material (biopsy or resection) was the diagnostic gold standard. All FNAs with subsequent surgical specimens were reviewed and assessed for cellularity, architectural features, and nuclear features. RESULTS: In 15 years, 5101 FNAs of the upper GI tract, hepatobiliary tract, and pancreas were performed. SCN was diagnosed in 98 (2%) patients. Seventy-two patients had definitive pathologic diagnoses: 68 were neoplastic and four were non-neoplastic. Cytomorphologic review in relationship to final diagnosis revealed three statistically significant features: low cellularity favors a benign process (P = .00544), epithelioid nuclear morphology favors malignancy (P = .00278), and identification of perinuclear vacuoles favors a diagnosis of GIST over non-GIST SCN (P = .04236). CONCLUSIONS: Among cases with follow-up, final pathologic diagnoses were SCN in 94% of cases diagnosed as SCN on FNA of upper GI, hepatobiliary tract, and pancreas. Although some cytomorphologic criteria are more suggestive of malignancy, arriving at a specific diagnosis relies on collaboration of clinical, radiologic, cytomorphologic, and immunohistochemical data.

Hybrid Capture 2 human papilloma virus testing for head and neck cytology specimens.

INTRODUCTION: High-risk human papilloma virus (hrHPV)-associated head and neck (HN) squamous cell carcinomas (SCCs) have important differences from non-hrHPV-related HNSCCs. A highly sensitive and specific test for HPV in cytology fine-needle aspirations (FNAs) would be useful, as it has the potential to alter therapy. MATERIALS AND METHODS: Patients with an HN FNA diagnosed as SCC or suspicious for SCC were included. Hybrid Capture 2 (HC2) was performed on residual rinse material and chromogenic in situ hybridization (CISH) for hrHPV was performed on the cell block. HC2-positive samples were genotyped for HPV types 16, 18, and 45. "Gold standard" p16 and CISH testing was performed on histologic material from the primary tumor. Tumors concordantly positive for p16 and CISH were considered hrHPV-positive, concordantly negative were considered hrHPV-negative, and discordant results were considered hrHPV-equivocal. RESULTS: A total of 96 FNAs from 95 patients were included. Surgical material was available in 80 patients. Of those, 29 patients (36%) were positive for hrHPV by "gold standard" testing, and 3 patients (4%) had equivocal results. HC2 was 72% sensitive and 100% specific for hrHPV. Sixty percent of HC2-positive aspirate samples were positive for HPV16. CISH was 61% sensitive and 79% specific for hrHPV. HC2 had a significantly better sensitivity and specificity than did CISH on paired sample analysis (P < .05). CONCLUSIONS: HC2 is a highly sensitive and specific assay for the detection of hrHPV in HN FNA samples. This new application of a familiar, widely available testing method has the potential to be clinically useful in the management of patients with HNSCC.