[1229] The Use of Progesterone Receptor (PR) and p53 in Assessing Uterine Smooth Muscle Tumors of Uncertain Malignant Potential (STUMP)

John Ross, Shi Wei, Kui Zhang, Michael Conner. University of Alabama at Birmingham, Birmingham, AL

Background: Uterine smooth muscle tumors (USMT) are extremely common, of which the overwhelming majority are benign leiomyomas (BLM) which pose little diagnostic difficulty. Similarly, as leiomyosarcomas (LMS) typically exhibit frankly malignant features including significant nuclear pleomorphism, tumor necrosis and brisk mitotic activity, the diagnosis is usually not overly challenging. However, a small subset of USMT cannot be diagnosed unequivocally as BLM or LMS, thus are classified as smooth muscle tumors of uncertain malignant potential (STUMP). Yet, the diagnosis of STUMP is largely subjective, with no generally accepted criteria. Because of this, attempts to further characterize USMT, specifically STUMP, are warranted. A recent paper by Hewedi, et al showed promise in this regard. The authors used a simple immunohistochemical (IHC) panel (PR and p53) and showed that all LMS show low PR expression and high p53 expression while all non-LMS (STUMP and BLM) stain oppositely (that is, low p53 and high PR).
Design: We sought to assess the usefulness of this panel (p53 and PR) in our cohort of USMT. A total of 27 USMT cases, including 5 BLM, 10 STUMP and 12 LMS, were included. Both were scored on a scale from 0-3 (0=negative, 3=strong, diffuse). We plan to correlate this with clinical follow up data (IRB approved; work in progress).
Results: While PR and p53 did separate many tumors into LMS or non-LMS (BLM/STUMP), this was far from 100%. Based on expected findings (as outlined above - Hewedi, et al), only 17 of 27 tumors behaved exactly as expected.

Combined PR and p53 Profile
Tumor TypeHigh PR/Low p53Low PR/High p53High PR/High p53Low PR/Low p53
Non-LMS (BLM/STUMP)13101

The clinical follow up in these cases will be key, and preliminary data reveal that outcomes generally correlate well with the H&E diagnosis. For example, even the LMS cases that stained unexpectedly (that is, high PR and low p53) had poor outcomes. Classification of LMS based solely on p53 and PR (classified as LMS if either p53 was 2 or 3 or PR was 0 or 1) gives a sensitivity of 0.667, a specificity of 0.867 and an accuracy of 0.778.

Classification of USMT based on p53 and PR

Conclusions: While H&E assessment remains the most important tool in diagnosing USMT, PR and p53 may serve as a useful adjunct in the proper classification of equivocal lesions, specifically STUMP. However, relying too heavily on IHC could be problematic. Correlation of these markers with clinical outcomes in these patients is ongoing.
Category: Gynecologic & Obstetrics

Wednesday, March 6, 2013 1:00 PM

Poster Session VI # 198, Wednesday Afternoon


Close Window