"Low-grade leiomyosarcoma" and late-recurring smooth muscle tumors of the uterus: a heterogenous collection of frequently misdiagnosed tumors associated with an overall favorable prognosis relative to conventional uterine leiomyosarcomas.

In most instances, uterine smooth muscle tumors (USMTs) are readily diagnosed as either benign or malignant. Rare patients whose smooth muscle tumors fail to meet leiomyosarcoma (LMS) diagnostic criteria will experience recurrence, and occasional cases of LMS patients experience a protracted clinical disease course. The aim of this study was to investigate whether "low-grade uterine LMS" can be defined as a clinicopathological entity and to learn which histologic features of USMTs correlate with indolent prognosis. We searched institutional databases for cases diagnosed between 1982 and 2008 that had been coded as low-grade LMS and/or cases coded as LMS that were associated with recurrences after 5 years of diagnosis. There were 185 cases with available clinical follow-up data (mean follow-up for survivors was 5.4 y); 57% of patients were dead of disease (DOD), 16% of patients were alive with disease (AWD), and 24% had no evidence of disease (NED). All available slides were reviewed by 2 pathologists (E.V. and R.S.) using Stanford USMT criteria to identify cases of bona fide LMS. Cases were not excluded if they failed to meet these criteria. Nine percent (16 of 185) of tumors had been coded as low-grade LMS. On review of cases with available slides (n=16), only 4 cases (25%) met Stanford USMT criteria for LMS, and 1 additional case was a myxoid LMS. Three cases were reclassified as endometrial stromal sarcomas with smooth muscle differentiation, and 7 cases (44%) failed to meet criteria for sarcoma [ie, they were atypical smooth muscle neoplasms (ASMNs)]. Six of 16 (38%) patients were NED with a mean follow-up of 76 months; 4 of 16 (25%) patients were AWD with a mean follow-up of 102.5 months; 4 of 16 (25%) patients were DOD with a mean follow-up of 79.2 months; and 2 of 16 (12.5%) patients died of unknown causes at 104 and 120 months. Despite being coded as having low-grade LMS in the database, none of the ASMN patients died of disease. Twelve percent of all cases (n=22) were associated with late recurrences. Of the 9 cases with available slides, 5 were bona fide LMSs, and 4 were ASMNs (coded as low-grade LMS in the database and included as part of the low-grade LMS portion of this study). Five of 9 (56%) patients were NED with a mean follow-up of 214 months; 2 (22%) are AWD with a follow-up of 107 and 201 months; and 2 patients were DOD with a follow-up of 108 and 143 months. Bona fide LMS recurrences in this group were earlier (mean, 6.2 y) and frequently fatal (2 of 5), whereas those patients with recurrent ASMN experienced disease progression later (mean, 12 y), and none died of disease. Whether "low-grade" uterine LMSs indeed exist is a matter still open for debate; however, when Stanford criteria are strictly applied, all tumors classified as LMSs should be regarded as intrinsically "high grade."