Serum protein acidic and rich in cysteine (SPARC) as a prognostic marker in soft tissue sarcomas

Clin Sarcoma Res. 2014 Jan 31;4(1):2. doi: 10.1186/2045-3329-4-2.

Abstract

Background: Serum protein acidic and rich in cysteine (SPARC) is a matricellular secreted glycoprotein that performs several cellular functions and has been implicated in tumorigenesis in a variety of tumor types. The chemotherapeutic agent nanoparticle albumin-encapsulated (NAB)-paclitaxel has been postulated to exploit SPARC expression to target neoplastic cells. SPARC's role, and potentially the role of NAB-paclitaxel, in the highly heterogeneous class of soft-tissue sarcomas (STS) has not been investigated. Our objective was to explore the pattern of SPARC expression and its prognostic significance in STS.

Methods: 27 tissue specimens representing various STS histologies were stained for SPARC expression by immunohistochemistry (IHC). Staining intensity was scored blindly. Survival was determined from patients' medical records and analyzed using Kaplan-Meier and log-rank with respect to SPARC expression level.

Results: Elevated SPARC expression was observed in 15/27 (56%) specimens. Overall patient survival segregated strongly based on levels of SPARC expression. Patients who expressed low-to-moderate levels of SPARC exhibited median survival of 22.1 months, while the median survival of patients with moderate-to-high expression levels was 4.4 months (log rank; p = 0.0016).

Conclusions: SPARC expression is elevated in a significant proportion of STS specimens analyzed in this study, but it does not appear to correlate with specific STS histologies. Given our limited sample size, we cannot draw definitive conclusions regarding association of SPARC with STS subtype. Overall survival segregates strongly by degree of SPARC expression, with elevated expression being adverse. If validated in a larger study, our results suggest that trials in STS with agents potentially targeting SPARC, such as NAB-paclitaxel, should be stratified by SPARC expression level.