• Gastrointestinal Cancers
• Oncology Clinical Trials

• Gastrointestinal Cancers
• Oncology Clinical Trials

Secreted protein acidic and rich in cysteine (SPARC) protein expression levels were not associated with overall survival (OS) or progression-free survival (PFS) in patients with metastatic pancreatic cancer from the Phase 3 Study of ABI-007 (Albumin-Bound Paclitaxel) plus Gemcitabine Versus Gemcitabine in Metastatic Adenocarcinoma of the Pancreas trial [MPACT; Hidalgo M et al. Ann Oncol 2014 (abstr O-0003)]. Manuel Hidalgo, MD, PhD, Centro Nacional de Investigaciones Oncológicas, Madrid, Spain, presented data from a sub-analysis of the MPACT trial.

Previous studies have suggested that increased expression of SPARC is correlated with decreased OS in patients with resectable pancreatic cancer [Infante JR et al. J Clin Oncol 2007]. In addition, a Phase 1/2 trial demonstrated that lower levels of SPARC were significantly associated with prolonged OS (p = .043) [Von Hoff DD et al. J Clin Oncol 2011]. The purpose of this sub-analysis of the MPACT trial was to further evaluate the relationship between SPARC and metastatic pancreatic cancer outcomes.

For this analysis, stromal fibroblasts and tumor epithelia harvested from predominantly metastatic lesion were assessed for SPARC levels using immunohistochemistry (IHC) with an anti-SPARC monoclonal antibody that was scored by 2 blinded pathologists [Hidalgo M et al. Ann Oncol 2014 (abstr O-0003)]. Stromal SPARC levels were considered high if ≥ 50% of fibroblasts stained positive. Tumor SPARC was measured assessing the histoscore, a well-established method for scoring protein expression in tissue that has heterogeneous staining in cell membranes, cytoplasm, and cell nuclei. Tumor SPARC was considered high if the histoscore was ≥ 100 and negative if the histoscore was 0. Enzyme-linked immunosorbent assay was used to evaluate SPARC levels in plasma collected at baseline and every 8 weeks in the MPACT trial. The IHC assay demonstrated 86% concordance between the Phase 1/2 and MPACT trials, and stromal SPARC was evaluable in 30% of patients from the MPACT trial.

Stromal SPARC expression was high in 71 out of 256 samples and was not associated with OS (HR, 1.019; p .903). In addition, tumor epithelial SPARC, which was low or negative in most samples, was also not associated with OS. Evaluable in 40% of patients, plasma SPARC levels were not significantly different between baseline and time points, and they were no associated with OS. PFS was not associated with SPARC expression levels in any of the samples.

In conclusion, Prof. Hidalgo indicated that, in his opinion, the data from this analysis of the MPACT trial suggest that SPARC expression was not prognostic for OS and was not predictive of treatment response in patients with metastatic pancreatic cancer. Therefore, SARC analysis requires further study and is not yet recommended to be used for treatment decisions in patient population.

• © 2014 MD Conference Express®