Predictors of Local Recurrence in Patients With Myxofibrosarcoma. (September 2018)

• Record Type: Journal Article
• Authors: Odei, Bismarck; Rwigema, Jean-Claude; Eilber, Frederick R.; Eilber, Fritz C.; Selch, Michael; Singh, Arun; Chmielowski, Bartosz; Nelson, Scott D.; Wang, Pin-Chieh; Steinberg, Michael; Kamrava, Mitchell
• Is Part Of: American journal of clinical oncology. Volume 41:Number 9(2018)
• Publication Date: 2018-09
• Subjects: sarcoma -- soft tissue sarcomas (STS) -- myxofibrosarcoma -- myxoid malignant fibrous histiocytoma (MFH) -- local recurrence of sarcoma; Cancer -- Treatment -- Periodicals; Oncology -- Periodicals; Tumors -- Periodicals; 616.994005
• DOI: 10.1097/COC.0000000000000382
• Languages: English
• ISSNs: 0277-3732

Abstract

Abstract : Objectives: Myxofibrosarcoma (MFS) is reported to have a higher risk of local recurrence (LR) following definitive surgical excision relative to other soft tissue sarcomas. We reviewed our clinical experience treating MFS to investigate predictors of LR. Materials and Methods: We retrospectively reviewed treatment outcomes for MFS patients treated at our institution between 1999 and 2015. A total of 52 patients were identified. Median age was 65 years (range, 21 to 86 y). Site of disease was: upper extremity (27%), lower extremity (46%), trunk (15%), pelvic (8%), and head and neck (4%). Patients had low, intermediate, high-grade, and unknown grade in: 23%, 8%, 67%, and 2% of tumors, respectively. Tumors were categorized as ⩽5 cm (35%), >5 cm (56%), or unknown size (9%). In total, 71% received radiotherapy: 19% preoperative, 50% postoperative, and 2% both. All patients underwent surgery. Margins were negative in 71%, close/positive in 21%, and unknown in 8%. In total, 27% of patients received chemotherapy. Univariate Cox regression analysis was utilized to determine associations between clinical and treatment factors with LR. Results: Median follow-up time was 2.9 years (range, 0.4 to 14.3 y). The 3-year actuarial LR, distant metastasis, and overall survival were: 31%, 15%, and 87%, respectively. Predictors of LR were patient age greater than or equal to the median of 65 years (hazard ratio, 13.46, 95% confidence interval, 1.71-106.18, P =0.013), and having close/positive tumor margins (hazard ratio, 3.4, 95% confidence interval, 1-11.53, P =0.049). Conclusions: In this institutional series of MFS older age and positive/close margins were significantly associated with a higher risk of LR.