Pancreatic cancer is a complex and heterogeneous disease that often lacks disease-specific symptoms in early stages. The malignancy is currently the fourth leading cause of cancer-related death in Western countries. In advanced stages, the overall 5-year survival is less than 1% to 2%. Most available treatments lack convincing cost-efficiency determinations and are generally not associated with relevant success rates. Targeting stromal components and stromal depletion is currently becoming an area of extensive research in pancreatic cancer. In this context, a glycoprotein, SPARC (secreted protein acidic and rich in cysteine) appears to play a central role. Still, the role of SPARC in carcinogenesis is controversial because conflicting results have been reported, and the pathways involved in SPARC signaling are not well established. Nonetheless, SPARC is highly expressed in the tumor stroma, principally in peritumoral fibroblasts, and the overexpression of SPARC in this compartment is associated with poorer prognosis. Interestingly, it has been suggested that SPARC present in the tumor stroma could sequester albumin-bound paclitaxel, enhancing the delivery of paclitaxel into the tumor microenvironment. In the present review, we summarize the known associations between SPARC and pancreatic cancer. Moreover, present and future therapies comprising SPARC-targeting are discussed.