Abstract P6-04-08: Cytosolic phospholipase A2 (cPLA2) as a therapeutic target in basal-like breast cancer

Abstract

Abstract Introduction: Basal-like breast cancer is frequently associated with triple negative phenotype, and there is a need for novel therapeutic strategies for this patient population. Inhibitors of cytosolic phospholipase A2 (cPLA2) have been suggested to block both MAPK and PI3K signalling and have a high potential for activity in basal-like breast cancer (Lin 1993, Wen 2013). In this study, we compared the expression of PLA2G4A between luminal B and basal-like breast cancer, both in patient-derived xenograft models (PDX) and human cancer tissue. In addition, we studied the effect of the novel cPLA2 inhibitor AVX235 on tumor growth in a basal-like PDX model. Materials and methods: Tumor tissue specimens were obtained from PDX models (n = 26) and a clinical breast cancer biobank (n = 32). Gene expression analysis was carried out on Agilent 8×60K microarrays. The expression of 54 genes directly involved in choline metabolism was examined. Differential expression of choline genes between basal-like and luminal B tumors was calculated by subtraction of log2 expression values. Mice carrying bilateral MAS98.12 basal-like xenografts (Bergamaschi 2009) were treated with the cPLA2 inhibitor AVX235 (30 mg/kg i.p. daily for 7 days then every second day for 14 days, n = 6) or drug-free vehicle (control group, n = 6). Results: The PDX models were subtyped into basal-like (n = 19) and luminal B (n = 7) subtypes based on the expression of 500 intrinsic genes [Sørlie 2003]. For the human cancer tissue, there were 18 basal-like and 14 luminal B tumors. There was a significant correlation between differential choline gene expression in basal-like vs luminal B tumors in PDX models and human cancer tissue (p<1.3*10−12). Both in PDX models (p<0.04) and human cancer tissue (p<0.0003), PLA2G4A was significantly higher expressed in basal-like than luminal B tumors. Treatment with AVX235 markedly reduced the growth rate of MAS98.12 xenografts compared to controls. After 19 treatmment days, the mean tumor volume (normalised to volume at start of treatment) in the treatment group was 36% of the tumor volume in the control group, the difference being statistically significant (p = 0.023). No signs of treatment-related adverse effects were observed. Conclusion: PLA2G4A is higher expressed in basal-like than in luminal B breast cancer. Treatment with the cPLA2 inhibitor AVX235 significantly inhibits tumor growth. These data suggest that cPLA2 inhibitors may be of particular value in treatment of basal-like breast cancer.