RESUMO
OBJECTIVE: There are no evidence based guidelines for the diagnosis of epithelial ovarian cancer (EOC) prior to the initiation of neoadjuvant chemotherapy. The aim of this study was to review our diagnostic practice, provide guidelines for clinical practice, and suggest a diagnostic strategy for validation in future prospective trials. PATIENTS AND METHODS: A retrospective chart review of patients undergoing neoadjuvant chemotherapy followed by debulking surgery was performed. Final diagnoses were based on expert pathology review of surgical specimens. Diagnostic strategies were defined as histologic, cytologic and clinical. Performance of these strategies in predicting final pathology was compared. RESULTS: Between 1994 and 2007, 149 patients were identified. Initial diagnosis was made on the basis of: cytology (paracentesis, thoracentesis, or fine needle aspirate) 72% (108 patients); histology (core biopsy, surgery) 18% (26), clinical (Radiology and CA-125) 10% (15). The final diagnosis was consistent with invasive EOC in 96% of patients. The diagnostic accuracies of the 3 strategies were: cytology 98%, histology 92%, and clinical 87%, (p=0.04). Specific EOC subtype was identified in 59% of patients (histology 77% and cytology 55%). When available, the initial subtype corresponded to the final subtype in 85% of cases: histology 80%, cytology 86%. CONCLUSION: Diagnosis of epithelial ovarian cancer based on cytology and histology are superior to clinical factors alone. In a centre with trained gynecologic cytopathologists, a diagnosis of EOC by cytology is not significantly inferior to a diagnosis made by histology. These data are important for clinical practice and the design of future clinical trials.