Abstract
According to recent reports, ovarian serous borderline tumor (SBT) harboring the BRAF V600E mutation is associated with a lower risk of progression to low-grade serous carcinoma. Preliminary observations suggest that there may be an association between eosinophilic cells (ECs) and the above-mentioned mutation, so this study aimed to evaluate interobserver reproducibility for assessing ECs. Forty-two samples of SBTs were analyzed for ECs with abundant eosinophilic cytoplasm. Immunohistochemical staining and genetic pro-filing were performed in all cases to verify the BRAF V600E mutation. A BRAF V600E mutation was found in 19 of 42 (45%) cases. Inter-observer reproducibility in the assessment of ECs was substantial (κ = 0.7). The sensitivity and specificity for predicting the mutation were 79% and 91%, respectively. Patients with BRAF-mutated SBTs were significantly younger than those without mutation (p = 0.005). SBTs with BRAF mutation were less likely to be accompanied by non-invasive implants than wild-type SBT: 12% (2/17) versus 33% (6/18). Seven cases were excluded due to incomplete cytoreductive surgery. Nevertheless, Fisher's exact test showed no significant differences between the two groups (p = 0.228). Overall, this study strengthens the idea that ECs in ovarian SBTs may represent a mutation with prognostic significance, which can serve as a primary screening test for BRAF V600E mutation in this pathologic entity.