Background/Objectives: Since non-invasive implants and invasive implants (metastases) are a key point of differentiation between serous borderline tumors (SBTs) and low-grade serous carcinoma (LGSC), the correct diagnosis of these two types of extraovarian lesions is crucial for patient treatment and prognosis. However, accurate diagnosis can be challenging even for experienced pathologists. The aim of this study was to evaluate interobserver agreement in the classification of these extraovarian lesions. Methods: Twenty-four cases of ovarian SBT and LGSC with 33 samples of non-invasive implants of SBT and metastasis of LGSC were independently reviewed by three gynecologic pathologists and three general pathologists. Diagnostic criteria included destructive invasion, micropapillary architecture, and retraction clefts. To measure interobserver agreement, Fleiss’ kappa and Cohen’s kappa were calculated, with consensus diagnoses determined by the majority of gynecologic pathologists. Results: According to the consensus, diagnosis 42.4% biopsies were classified as metastases of LGSC and 57.6% as non-invasive implants of SBT. Overall reproducibility was substantial (κ = 0.61). The agreement among gynecologic pathologists, as well as between gynecologic pathologists and the consensus (using leave-one-out reference), was substantial to near-perfect (κ = 0.745–0.821). General pathologists’ agreement with the consensus was moderate (κ = 0.467–0.698). Agreement between general pathologists was also moderate, with κ values ranging from 0.413 to 0.518. The difference in pairwise agreement between the two groups was statistically significant, confirming that gynecologic pathologists outperformed general pathologists in classifying extraovarian lesions. Conclusions: The results showed that current diagnostic reproducibility remains suboptimal, particularly among general pathologists, underscoring the need for improved training and standardized criteria. Ultimately, a multidisciplinary approach combining morphological expertise, immunohistochemical validation and molecular stratification will be essential for optimizing diagnosis and treatment.
Badlaeva et al. (2026) studied this question.