ORIGINAL ARTICLE Multifocality in Retroperitoneal Sarcoma A Prognostic Factor Critical to Surgical Decision-Making Daniel A. Anaya, MD,* Guy Lahat, MD,* Jun Liu, MS,† Yan Xing, MD, MS,* Janice N. Cormier, MD, MPH,*† Peter W. Pisters, MD,* Dina C. Lev, MD,‡ and Raphael E. Pollock, MD, PhD* Objective: To evaluate the significance of multifocality on overall survival (OS) in patients with retroperitoneal sarcoma (RPS) and establish a data- derived, prognostically and therapeutically useful definition of sarcomatosis. Summary Background Data: The incidence, clinical features, and prog- nostic significance of multifocality in RPS is unknown. No current standard- ized definition for sarcomatosis is available. Methods: We conducted a retrospective analysis of 393 patients with primary or recurrent nonmetastatic RPS treated at a comprehensive cancer center between 1996 and 2006. Baseline and treatment variables were compared in patients with unifocal and multifocal disease. A multivariate model was used to evaluate the association of multifocality and OS and identify additional prognostic factors in patients with multifocal disease. Results: The median follow-up time for all patients was 69 months; 79 patients (20%) presented with multifocal disease. The 5-year OS rate was less in the multifocal group compared with the unifocal group (31% vs. 60%, respectively; P  0.0001). After multivariate analysis, multifocality re- mained an independent predictor of worse OS {hazard ratio (HR) 1.7 (95% confidence interval (CI), 1.2–2.5); P = 0.004}. Additionally, patients with more tumors had significantly worse prognosis (7 tumors, HR 2.1 (95% CI, 1.1–3.9); P = 0.03), with a 5-year OS rate of 7%. Conclusions: Multifocal RPS is associated with worse OS in patients with either primary or recurrent RPS; Patients with 7 tumors had the worst prognosis. This criterion can be used to define sarcomatosis, thereby iden- tifying patients whose survival will ultimately depend on effective systemic therapy. (Ann Surg 2009;249: 137–142) R etroperitoneal sarcoma (RPS) comprises 10% of soft tissue sarcoma (STS), with an incidence of approximately 1000 new cases diagnosed every year in the United States. This incidence has remained stable over the last 3 decades, 1,2 corresponding to 2.7 cases per million diagnosed annually. 3 RPS is characterized by heterogeneous tumors, with a vast array of histologic subtypes and varying presenting clinical characteristics. RPS is also characterized by aggressive behaviors with high local recurrence rates (70%, 5-year local recurrence rate) and modest survival rates (50%, 5-year overall survival (OS)). 4,5 In contrast to extremity STS, and despite the introduction of new chemotherapeutic agents and mod- ern radiation techniques, multimodality therapy has not significantly improved outcomes in patients with RPS, and surgery remains the mainstay of treatment. 6–8 The rarity of RPS and the array of presenting histologic subtypes have significantly constrained study of these tumors. Most published series include a small number of patients who have heterogeneous clinical and histopathologic characteristics. Although data are limited, several independent predictors of recurrence and survival have consistently been identified for patients with RPS, including type of presentation (primary versus recurrent), complete- ness of resection, histology, and tumor grade. 5,9 –13 Multifocality in STS, defined as having more than 1 noncon- tiguous tumor, has traditionally been considered a feature of more aggressive disease associated with worse long-term outcomes com- pared with unifocal STS. However, this finding is derived from studies evaluating patients with extremity STS. 14,15 The incidence and clinicopathologic characteristics of multifocality in patients with RPS has not been well characterized, and its prognostic significance is unknown. Furthermore, the term “sarcomatosis” is often used in the context of advanced RPS, implying poor overall prognosis. However, because there is currently no standard definition, the term sarcomatosis per se does not provide bonafide meaningful or objec- tive prognostic information. The goal of this study was to evaluate the incidence, clinico- pathologic features, and prognostic significance denominated by multifocality using a large statistically robust cohort of patients treated for nonmetastatic primary or recurrent RPS. Our second goal was to establish a prognostic-based definition of sarcomatosis that would identify a highest-risk patient subset of those who presented with multifocal disease. PATIENTS AND METHODS Patient Selection This study was approved by the institutional review board of University of Texas MD Anderson Cancer Center. Using the STS database at MD Anderson Cancer Center, we identified 1120 patients who were evaluated for RPS between 1996 and 2006, exclusive of desmoid or gastrointestinal stromal tumors (GIST). We then selected patients without distant metastatic disease, which yielded 393 patients with primary or recurrent RPS. We reviewed these patients’ medical records and retrieved informa- tion about their demographic and clinicopathologic characteris- tics and treatment and outcome variables. All 393 patients were prospectively evaluated by a multidisciplinary sarcoma team at MD Anderson before initiation of therapy, and surgical resection was of- fered when feasible. Patients were considered to have unresectable disease when there was proximal encasement of the mesenteric vessels or when the patient declined surgical treatment in the basis of extensive surgery required to achieve complete resection (ie, hemipelvectomy). In addition, adjuvant therapies, including primar- ily doxorubicin-based chemotherapy and also radiation therapy, were selectively employed as deemed appropriate by the treating physicians. From the Departments of *Surgical Oncology, †Biostatistics and Applied Math- ematics, and ‡Cancer Biology, MD Anderson Cancer Center, University of Texas, Houston, Texas. Supported in part by AJCC grant awarded to Raphael E. Pollock towards development of a soft tissue sarcoma staging system. Reprints: Raphael E. Pollock, MD, PhD, Department of Surgical Oncology, Unit 444, University of Texas MD Anderson Cancer Center, 1400 Holcombe Blvd, PO Box 301402, Houston, TX 77030-4009. E-mail: rpollock@mdanderson.org. Copyright © 2009 by Lippincott Williams & Wilkins ISSN: 0003-4932/09/24901-0137 DOI: 10.1097/SLA.0b013e3181928f2f Annals of Surgery • Volume 249, Number 1, January 2009 137