Research Abstract

One of the major obstacles to effectively treating cancer is the development of drug resistance, due to the tremendous genomic instability present in virtually all tumors. In addition to resistance to therapy, tumors are also to evade the immune system by presenting “self” makers on the cell surface. One example of such a “self” marker is CD47, which binds SIRPα on macrophages and prevents phagocytosis, essentially acting as a “do not eat me” signal. CD47 is also a receptor for the anti-tumorigenic protein thrombospondin-1 (Tsp-1) and binding of Tsp-1 to CD47 blocks the interaction with SIRPα and promotes macrophage-mediated cell killing. In fact, Tsp-1 expression is normally down-regulated in tumor cells via activation of signaling pathways or constitutive activation of transcription factors and its down-regulation has been shown to be a mechanism of acquiring resistance to taxanes. However, these pathways are generally not mutated in the cells that comprise the tumor microenvironment. This presents a novel opportunity to attack the tumor by targeting the microenvironment.

Through a functional proteomic screen we have identified tumor-secreted proteins that modulate Tsp-1 expression in the tumor microenvironment. Specifically, the cells in which Tsp-1 is most strongly modulated are bone marrow derived myeloid suppressor cells (MDSCs) and macrophages. These cells are normally coopted by the tumor to stimulate growth and metastasis. In fact, through our screen, we have identified a protein, PRSS2, secreted by highly metastatic breast and prostate cancer cells that represses Tsp-1 expression in the microenvironment, and are in the process of developing a therapeutic antibody to block its activity.

Moreover, we have identified a protein, prosaposin, which stimulates Tsp-1 expression and converts these bone marrow derived cells into anti-tumorigenic agents. Based on these findings we developed a peptide derived from prosaposin that retains the Tsp-1 stimulating activity of the full-length protein. We have demonstrated that the prosaposin peptide is able to regress tumor growth in animal models at both the primary and metastatic sites when administered in a systemic manner. As a result of these findings, we are in the process of developing the prosaposin peptide as a therapeutic agent to treat multiple types of cancer, based on the expression of CD47.

Strikingly, we have found, through a multivariate analysis of over 250 prostate and endometrial cancer patients that prosaposin expression correlates with patient survival better than any other currently used prognostic marker. As such, we examined whether prosaposin expression, or the lack thereof, mediated drug resistance. To that end we examined multiple matched drug-sensitive (parental) and drug-resistant tumor cell lines and found that in all cases drug-resistant cells expressed significantly lower levels of prosaposin. Thus, repression of prosaposin could represent a novel method of acquiring drug resistance.