Abstract Introduction/Objective Epithelioid variant of the GIST can mimic carcinoma and misdiagnosis has grave implications in patient management. The purpose of this report is to describe such a case and to remind pathologists to keep epithelioid GIST in the differential. Methods/Case Report A 66-year-old male presented with melena for 2 weeks. Endoscopy showed a large, infiltrative, and ulcerated mass in the gastric fundus. CT imaging showed a 16 x 12 cm mass-like gastric wall thickening, most compatible with gastric carcinoma. The mass was biopsied, which showed sheets and clusters of anaplastic, large to medium sized polygonal epithelioid cells with areas of rhabdoid phenotype characteristic of an undifferentiated gastric carcinoma. Initial immunohistochemical stains including pancytokeratin and p40 showed negativity, prompting additional stains including Cam5.2, SOX10, CD45, synaptophysin, INSM1, EMA, CK7 and CK20- all of which were also negative. At this point the main diagnostic consideration was undifferentiated gastric carcinoma. However, for completion purposes, DOG1 stain was added that came back positive. Since rhabdoid morphology is a bit unusual, even for epithelioid GISTs, and literature review suggested that DOG1 positivity can be seen in up to 11% of carcinomas, confirmatory NGS testing was performed. This demonstrated a KIT W557R gene mutation, confirming the diagnosis of GIST. The patient was treated with imatinib and subsequent imaging showed decreased size and hypermetabolism of the gastric mass. Patient is alive with disease 6 months after the diagnosis. Results (if a Case Study enter NA) NA Conclusion Using a panel of appropriate immunohistochemical stains and keeping epithelioid GIST in the differential can help avoid this diagnostic pitfall.
Read full abstract