Complex Atypical Hyperplasia: Improving the Prediction of Associated Carcinoma
MM Leitao, Jr, G Han, NR Abu-Rustum, RR Barakat, RA Soslow. Memorial Sloan-Kettering Cancer Center, New York
Background: The primary objective was to identify factors associated with an endometrial carcinoma (CA) diagnosis on hysterctomy after a diagnosis of complex atypical hyperplasia (CAH) on biopsy or curettage.
Design: We retrospectively identified all CAH cases diagnosed on biopsy (BX) or curettage (EMC) from March 1994-May 2008 with follow up hysterectomies. CAH cases were subclassified as: CAH-suspicious if features bordered on carcinoma or carcinoma could not be excluded; CAH-polypoid if CAH arose in or was associated with a polyp; CAH-focal; or CAH-NOS (not otherwise specified). The categories were not mutually exclusive. A subset analysis was performed for cases diagnosed by a gynecologic pathologist (subspecialist CAH) to determine whether any differences in CAH diagnostic criteria affected rates of CA in followup hysterectomy.
Results: We identified 197 CAH cases. The median age was 54 y (range 32-86 y). The median time from BX or EMC to hysterectomy was 47 days (5-572 days). CA was subsequently diagnosed on hysterectomy in 34% of cases. CA was diagnosed after: CAH-suspicious in 56% of cases, compared to 28% not suspicious; CAH-polypoid in 20% of cases, compared to 38% nonpolypoid; and CAH-focal in 19%, compared to 40% nonfocal. All comparisons were statistically significant (p-values=0.001, 0.02 and 0.005, respectively). Method of preoperative endometrial sampling, age, menopausal status and BMI were associated with CA on univariate analysis. On multivariate analysis, CAH-suspicious, CAH-nonfocal, CAH-nonpolypoid, BX compared to EMC, and older age were independently associated with an increased risk of CA (all p-values <0.001). 112 of the 197 cases were subspecialist CAH cases, and these were as likely to be followed by a CA diagnosis as other cases (34% versus 34%; p=NS). Subspecialist CAH cases were as frequently associated with myoinvasive CA as other cases (13% versus 15%; p=NS) and rates of deep myoinvasion (>50%) did not differ between groups (4% versus 5%). All hysterectomy diagnoses of CA were FIGO grade 1 with only rare exceptions, n=2 in the subspecialist CAH group and n=3 in the others.
Conclusions: The rate of CA on hysterectomy after a diagnosis of CAH is influenced by the method of sampling, age and CAH type. Subspecialty review of CAH cases by a gynecologic pathologist had no apparent impact on the ability to predict CA, including myoinvasive CA, on follow up.
Category: Gynecologic & Obstetrics
Wednesday, March 24, 2010 9:30 AM
Poster Session V # 123, Wednesday Morning