A prognostic model for resectable soft tissue and cutaneous angiosarcoma. Issue 5 (4th July 2016)

• Record Type: Journal Article
• Authors: Sinnamon, Andrew J.; Neuwirth, Madalyn G.; McMillan, Matthew T.; Ecker, Brett L.; Bartlett, Edmund K.; Zhang, Paul J.; Kelz, Rachel R.; Fraker, Douglas L.; Roses, Robert E.; Karakousis, Giorgos C.
• Is Part Of: Journal of surgical oncology. Volume 114:Issue 5(2016)
• Page Start: 557
• Page End: 563
• Publication Date: 2016-07-04
• Subjects: angiosarcoma -- prognosis -- risk factor -- survival; Cancer -- Surgery -- Periodicals; Neoplasms -- Periodicals; 616
• DOI: 10.1002/jso.24352
• Languages: English
• ISSNs: 0022-4790

Abstract

Abstract : Background: Angiosarcoma is an aggressive tumor rising in incidence from use of therapeutic radiation. Because of its relative rarity, prognostic factors have not been clearly delineated. Methods: Patients who underwent resection of localized angiosarcoma from 2002 to 2012 were identified using the National Cancer Data Base. Factors associated with poor overall survival (OS) were identified using Cox proportional hazards modeling and internally validated with bootstrap resampling. These were used to construct a risk model to identify low‐, intermediate‐, or high‐risk groups. Results: Median OS among 821 patients undergoing resection was 3.4 years. On multivariable analysis of factors known preoperatively, those associated with worse OS included: age >70 years (HR 2.02, P < 0.0001), black race (HR 1.92, P < 0.0001), head and neck primary (HR 1.44, P = 0.003), grade 3 tumor (HR 1.53, P = 0.013), size 3–7 cm (HR 1.64, P < 0.0001), size >7 cm (HR 2.37, P < 0.0001). After including postoperative variables, positive resection margins were associated with worse OS (microscopic, HR 1.59, P = 0.002; macroscopic, HR 3.38, P = 0.008). Stratification by risk group was superior to AJCC stage in discriminating OS. Conclusions: In the largest study to date of patients with angiosarcoma, risk factors for poor OS were identified to create a clinically useful risk model that can prognosticate patients with localized disease following surgical resection. J. Surg. Oncol. 2016;114:557–563 . © 2016 Wiley Periodicals, Inc.