[1157] Can We Accurately Predict High Risk Endometrial Carcinoma Preoperatively?

V Parkash, N Rassaei, GL Keeney, CN Otis, JL Hecht, O Fadare. Yale University, New Haven, CT; Mayo Clinic, Rochester, MN; Baystate Health Systems, Springfield, MA; Beth Israel Deaconess Medical Center, Boston, MA; Vanderbilt University, Nashville, TN

Background: The management of endometrial carcinoma (EMCA) requires accurate staging. Most patients present with stage 1 disease, but prediction of stage is inaccurate in 15-20% of cases when using clinical parameters and FS evaluation. Therefore, the NCCN guidelines recommend complete staging of ALL cases of EMCA. This, however, means that 15-20% of patients get staging for low stage, low grade disease, which may be associated with some morbidity. The Mayo Clinic has proposed criteria that may allow more accurate identification of such low risk patients with endometrial biopsy (EMB) and intraoperative assessments (IO), and avoid staging in these cases. This study was carried out to determine if these triage criteria worked at an institution other than the Mayo Clinic.
Design: Pathology reports on patients with EMCA, who had IO and staging were reviewed over a one year period and patients were classified as high risk based on EMB and IO. Patients were classified as high risk if they had any of the following -- grade 3 carcinoma on EMB or IO, tumor size > 2cm, myoinvasion of > 50% and lymphovascular invasion (LVSI) at IO. The final reports of the remainder of patients (putative “low risk”) were reviewed to determine if any “high risk cases” were inaccurately identified using this protocol.
Results: Of a total of 85 patients who underwent staging for EMCA, 52 patients had EMB's reported as low grade EMCA. Of these, 5 had ovarian enlargement that on frozen section (FS) revealed carcinoma (4 synchronous ovarian carcinoma). 23 patients had bulky endometrial tumors, > 2 cm grossly. Of the remaining 24 cases, 3 had depth of invasion > 50%, while1 had a high grade carcinoma at FS (a minimal serous carcinoma). All the remaining 20 cases remained low stage (16 stage 1A, of which 3 had only complex hyperplasia on hysterectomy). Of the four stage 1B cases, 1 had a single focus of LVSI. In this case, the depth of invasion was 30%.
Conclusions: Using a strategy of exclusion of patients with negative predictive parameters, “low risk patients” were identified accurately in > 95% of cases. Application of this strategy may save an estimated 20% of patients an expensive and potentially morbid staging procedure.
Category: Gynecologic & Obstetrics

Monday, March 22, 2010 1:00 PM

Platform Session: Section C, Monday Afternoon


Close Window