FAP expression in deep and superficial endometriosis: Association with MRI and transvaginal ultrasound visibility.

PURPOSE: Fibroblast activation protein (FAP) is expressed in the fibroinflammatory stroma of endometriosis, but its relationship to conventional imaging visibility is unknown. We evaluated FAP expression in deep endometriosis (DE) and superficial endometriosis (SE) and its relationship to lesion visibility on magnetic resonance imaging (MRI) and transvaginal ultrasound (TVUS). METHODS: We retrospectively analyzed specimens from DE (n = 13) and SE (n = 9) patients who underwent MRI and/or TVUS. FAP expression was assessed immunohistochemically using H-scores. Two abdominal radiologists independently graded the visibility of pathology-confirmed lesions on MRI and TVUS using a study-specific 0-3 scale, with N/A assigned when the lesion location was outside the available imaging field of view. H-scores were compared by phenotype and imaging visibility (lower [0-1] vs higher [2-3]) using nonparametric analyses. RESULTS: FAP expression did not differ significantly between DE and SE (median H-scores 170 vs 200, p= 0.738). Endometriosis showed substantial FAP expression, although H-scores were lower than colorectal adenocarcinoma controls (p < 0.05). On MRI, no lesions were assigned N/A; low-/high-visibility groups included 5/16 lesions for Reader 1 and 3/18 for Reader 2, with median H-scores of 220/170 for both readers (both p > 0.05). On TVUS, 1 and 2 lesions were assigned N/A for Readers 1 and 2, respectively; among evaluable lesions, low-/high-visibility groups included 2/10 and 1/10 lesions, with median H-scores of 48/165 and 20/160, respectively, limiting conclusions regarding visibility and FAP expression. CONCLUSION: DE and SE showed substantial stromal FAP expression, including SE lesions poorly visible on conventional imaging. These findings provide a histopathological rationale for FAP-targeted imaging in endometriosis.