Estrogen Receptor (ER) and Progesterone Receptor (PR) Immunohistochemistry is Sensitive and Specific for Differentiating Retroperitoneal Leiomyosarcomas With Symplastic-like Features From Their Uterine Mimics

Previous studies have evaluated the utility of estrogen receptor (ER) and progesterone receptor (PR) immunohistochemistry (IHC) in differentiating uterine versus extrauterine leiomyosarcomas (LMS). At best, these studies have shown only modest sensitivity and specificity for these markers in this context. In our own practice, we have noticed that retroperitoneal LMS, such as those arising in the wall of the inferior vena cava, frequently exhibit a remarkable resemblance not to uterine LMS, but rather to uterine leiomyomas (LM) with bizarre nuclei, formerly known as symplastic LM. This includes areas with bland nuclear cytology, punctuated by the presence of cells with large bizarre nuclei but a paradoxically low mitotic index. We refer to these areas in retroperitoneal LMS as "symplastic-like." It has been our experience that these "symplastic-like" areas are frequently the predominant or exclusive component in small core biopsies of retroperitoneal LMS, even when the resection of these tumors reveals the presence of more conventional high-grade LMS morphology. In female patients, symplastic-like morphology in a smooth muscle tumor at an intra-abdominal site raises the possibility of iatrogenic dissemination of a uterine LM with bizarre nuclei from a prior myomectomy or morcellation procedure. We hypothesized that negative staining for ER and PR by IHC could effectively exclude a uterine origin, given the high sensitivity of these markers for all variants of uterine LM. After successfully using ER and PR IHC in our clinical practice on a few index cases, we decided to study a larger cohort of carefully selected cases to systematically determine the sensitivity and specificity of these markers in this very specific context. Confining our search to include only female patients, we identified 8 cases of retroperitoneal LMS that had been confirmed radiologically, intraoperatively and/or histologically to originate from a retroperitoneal source and 6 cases of uterine-based LM with bizarre nuclei, all diagnosed at our institution over an 8-year period. We tested only whole slides for ER and PR IHC. ER and PR were both completely negative in all 8 cases of retroperitoneal LMS and were both strongly expressed in all 6 cases of LM with bizarre nuclei. In conclusion, despite conflicting data in the literature regarding the utility of ER and PR in distinguishing uterine versus extrauterine smooth muscle tumors, we endorse the use of these markers for the specific distinction of retroperitoneal LMS with symplastic-like features from disseminated uterine LM with bizarre nuclei in female patients.