Lack of uniformity in reporting autoimmune gastritis among a diverse group of pathologists.

Autoimmune gastritis (AIG) is a clinicopathologic diagnosis requiring characteristic histopathology and correlation with laboratory work-up. To better understand how the diagnosis of AIG is made and reported in the pathology community, we conducted an anonymous web-based survey which was circulated among a diverse group of pathologists. Excluding trainees there were 64 respondents: 25 academic gastrointestinal pathologists (AGI, 39%), 22 academic general pathologists (AGP, 34%), 17 private general pathologists (PP, 27%). Our survey results highlighted variations in work-up and sign-out practices. The type of metaplasia needed to diagnose AIG lacked consensus. There was variation in accurate interpretation of immunostains with a trend towards more accurate diagnosis of enterochromaffin-like (ECL) cell hyperplasia by AGI (92%) and AGP (95%) than PP (71%) (p = 0.07). G-cells in antrum on neuroendocrine immunostain, a mimicker of ECL cell hyperplasia, was more frequently misdiagnosed by PP/ AGP (44%), versus AGI (12%) (p = 0.02). A triple immunostain panel (H. pylori, neuroendocrine, gastrin) was used in the work-up of AIG by 72% of AGI versus 23% AGP and 12% PP (p = 0.000061). The less-specific term "atrophic gastritis" was used in the diagnostic line more by respondents with >10 years sign-out experience compared with others (p = 0.04). In conclusion, the survey results highlighted deficiencies in the interpretation of neuroendocrine immunostains which is crucial for AIG diagnosis, as well as variation in reporting practices and definitions. Uniform criteria and terminology are needed in this field to improve communication with clinicians, resulting in appropriate testing and follow-up.

Pediatric myeloid sarcoma: a single institution clinicopathologic and molecular analysis.

Myeloid sarcoma (MS) is a neoplastic condition composed of immature myeloid cells involving an extramedullary site. We investigated underlying chromosomal and molecular alterations to assess potential molecular markers of prognosis and outcome in this rare pediatric disease. We conducted a retrospective review of clinicopathologic and cytogenetic data from 33 pediatric patients with MS (ages 1 month-18 years) at our institution over a 32 year period (1984-2016). Tissue-based cancer microarray and targeted next-generation sequencing analysis were performed on six cases. The median age at diagnosis was 2.8 years with a male-to-female ratio of 2.6:1. MS is commonly presented with concomitant marrow involvement (n = 12, 36.4%) or as a recurrence of acute myeloid leukemia (AML; n = 14, 42.4%). The skin (n = 18, 54.5%) and soft tissue (n = 9, 27.3%) were the most common sites of involvement. Twenty-one of 25 samples (84.0%) harbored chromosomal aberrations; KMT2A alterations (n = 10, 40.0%) or complex cytogenetics (n = 7, 28.0%) were most frequent. Mutations in RAS, tyrosine kinase, cell signaling, and chromatin remodeling genes were detected. When compared to pediatric patients with AML without extramedullary involvement (EMI), inferior overall survival (OS) was observed (18.8 months vs. 89.3 months, p = .008). Pediatric patients with MS with non-favorable cytogenetics [abnormalities other than t(8;21), inv(16)/t(16;16), or t(15;17)] had a significantly lower OS compared to patients with AML with non-favorable cytogenetics and no extramedullary involvement (8.0 months vs. 28.1 months, p < .001). Pediatric MS is a rare disease with diverse clinical presentations. Non-favorable cytogenetics may be a poor prognostic marker for pediatric patients with MS and molecular diagnostics can assist with risk stratification and identify potentially actionable targets.