Immunohistochemistry for TFE3 lacks specificity and sensitivity in the diagnosis of TFE3-rearranged neoplasms: a comparative, 2-laboratory study

TFE3 rearrangements are characteristic of alveolar soft part sarcomas (ASPS), Xp11.2 translocation renal cell carcinomas (Xp11-RCC), and other rare tumors. Immunohistochemistry for TFE3 protein has been considered by some to be a reliable surrogate for TFE3 molecular studies, although others disagree. We compared 2 methods for TFE3 immunohistochemistry to determine if technical differences underlie these differences. Ninety-eight archival cases of mixed type, 19 ASPS, and 8 Xp11-RCC were stained for TFE3 at Laboratory A and Laboratory B using routine protocols. Positive controls were normal human testis (Laboratory A) and Xp11-RCC (Laboratory B). Nuclear staining was scored as "negative," "1+" (<10%), "2+" (10%-50%), and "3+" (>50%). Intensity was scored as "negative," "weak," "moderate," or "strong." Only moderate-strong, 2+ or 3+ staining was considered positive. Laboratory A results were as follows: archival cases (42 of 98, 43%), ASPS (16 of 19, 84%), and Xp11-RCC (7 of 8, 88%). Laboratory B results were as follows: archival cases (5 of 98, 5%), ASPS (14 of 19, 74%), and Xp11-RCC (5 of 8, 63%). TFE3 fluorescence in situ hybridization was positive in all tested ASPS and Xp11-RCC. The overall sensitivity and specificity of TFE3 immunohistochemistry for TFE3-rearranged neoplasms were 85% (23/27) and 57% (56/98) at Laboratory A and 70% (19/27) and 95% (93/98) at Laboratory B. Technical differences, in particular, the type of control tissue, likely account for these different results. The results of our study and prior studies suggest that TFE3 immunohistochemistry should play only a minor role (if any) in the diagnosis of TFE3-rearranged tumors, with fluorescence in situ hybridization representing the preferred method.