Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy with multiple genetic and epigenetic alterations and desmoplastic stroma, all of which contribute to therapy resistance. Standard-of-care chemotherapies only modestly benefit some patients in unselected populations, meaning more personalised treatments are desperately needed. Arginine methylation is a universal post-translational modification catalysed by protein arginine methyltransferases (PRMTs). Inhibiting PRMTs has emerged as a promising therapy in several cancers including PDAC, which overexpress PRMTs and in particular PRMT5 [1]. High expression of PRMT5 promotes cancer cell proliferation and is associated with poor prognosis in PDAC [2].
Patient-derived cell line models of PDAC were screened using cytotoxicity assays with various concentrations of SAM-competitive (LLY-283) and MTA-cooperative (MRTX-1719) PRMT5 inhibitor. A broad range of sensitivity to PRMT5 inhibition was observed across different cell lines and only MRTX-1719 sensitivity correlated with MTAP loss as previously reported [3]. To investigate other potential biomarkers of cancer cell sensitivity to PRMT5 inhibition, DNA methylation analysis was performed. Gene Ontology enrichment analysis in the PRMT5-treated sensitive lines demonstrated that genes with increased methylation were related to cell cycle G1/S transitions, while genes with reduced methylation were related to DNA repair processes. Therefore, we assessed synergy of PRMT5 inhibitors in combination with DNA damaging agents and other targeted therapies using Bliss independence synergy index (SI). We identified MAT2A and PARP inhibitors to be synergistic (SI>0) to additive (SI=0) with PRMT5 inhibitors in a range of PDAC models, both in vitro and in vivo, irrespective of MTAP loss or homologous recombination deficiency (a known predictor of PARP sensitivity). This suggests that PRMT5 sensitivity may go beyond MTAP status, with new biomarkers yet to be discovered. Moreover, using 3D tumour cell-cancer associated fibroblast (CAF) co-culture assays, we showed no direct effects of PRMT5 targeting on CAF contractility and matrix remodelling; however, inhibition of PRMT5 as a monotherapy or combination therapy with MAT2A or PARP inhibitor significantly reduced cancer cell invasion in 3D organotypics model.
Our ongoing work focuses on understanding the biological determinants, including gene expression and methylation signatures, and mechanisms behind PRMT5 inhibitor response and potential synergy that may allow broader clinical application.