Abstract
Background/Objectives: Gastrointestinal stromal tumors (GISTs) are primarily identified using KIT and DOG1 immunohistochemistry, although staining may be absent or heterogeneous in some cases. Cytoplasmic Wilms tumor 1 (WT1) expression has been reported in GISTs, but its immunohistochemical characteristics and distribution among other mesenchymal tumors remain incompletely characterized. This study aimed to evaluate cytoplasmic WT1 expression in GISTs, compare its staining profile with established GIST-associated markers, and examine its distribution in a non-GIST mesenchymal comparison cohort. Methods: Twenty-eight GISTs were evaluated for WT1, KIT, DOG1, CD34, and h-caldesmon expression. Staining intensity and extent were assessed using an immunoreactivity score (IRS; 0–12). The comparison cohort comprised 17 non-GIST mesenchymal lesions: eight leiomyomas, six schwannomas, two inflammatory fibroid polyps, and one desmoid-type fibromatosis. IRS distributions were analyzed using the Friedman test, followed by Wilcoxon signed-rank tests with Bonferroni correction. Kruskal–Wallis and Mann–Whitney U tests were used for comparisons within the non-GIST cohort. Results: All 28 GISTs exhibited strong, diffuse cytoplasmic WT1 immunoreactivity, with a maximal IRS of 12. In contrast, IRS values varied for KIT (median, 8; range, 0–12), DOG1 (median, 12; range, 4–12), CD34 (median, 12; range, 0–12), and h-caldesmon (median, 9; range, 0–12). Overall, IRS distributions differed significantly across the five markers (Friedman χ2(4) = 31.18, p < 0.001), with WT1 showing higher scores than each comparator marker in post-hoc analyses. Five GISTs were KIT-immunonegative, including three spindle-cell and two epithelioid tumors; all retained DOG1 expression and demonstrated maximal WT1 scores. Among non-GIST lesions, all six extra-gastrointestinal schwannomas exhibited maximal cytoplasmic WT1 immunoreactivity, whereas leiomyomas, inflammatory fibroid polyps, and desmoid-type fibromatosis were WT1-negative. WT1 scores were higher in GISTs than in the overall non-GIST cohort and the non-neural mesenchymal subgroup (both p < 0.001), whereas GISTs and schwannomas showed identical IRS distributions. Additional WT1 immunoreactivity was observed in myenteric plexus regions and vascular-associated structures, while striated muscle fibers exhibited a characteristic sarcomeric staining pattern. Conclusions: Strong, diffuse cytoplasmic WT1 immunoreactivity was consistently observed in this GIST series, including KIT-immunonegative and epithelioid tumors. However, comparable expression in extra-gastrointestinal schwannomas indicates that cytoplasmic WT1 staining is not specific for GIST. Its potential complementary diagnostic value requires further validation and should be interpreted alongside morphology and established immunohistochemical markers.