PRMT5 Inhibition Promotes FOXO1 Tumor Suppressor Activity to Drive a Pro-Apoptotic Program That Creates Vulnerability to Combination Treatment with Venetoclax in Mantle Cell Lymphoma

Mantle cell lymphoma (MCL) is an incurable B cell malignancy, comprising 5% of non-Hodgkin lymphomas diagnosed annually. MCL is associated with a poor prognosis due to emergence of resistance to immuno-chemotherapy and targeted agents. The average overall survival of patients with MCL is 4-6 years and for the majority of patients who progress on targeted agents, survival remains at a dismal 3-8 months. There is a major unmet need to identify new therapeutic approaches that are well tolerated to improve treatment outcomes and quality of life.

The type II protein arginine methyltransferase enzyme, PRMT5 is overexpressed and promotes growth and survival of MCL. Inhibition of PRMT5 with a novel, SAM-competitive class of inhibitors drives anti-tumor activity in MCL cell lines and patient derived xenograft (PDX) models derived from patients with relapse or refractory disease. Selective inhibition of PRMT5 with PRT-382 (Prelude Therapeutics) in these models and MCL cell lines leads to disruption of constitutive PI3K/AKT signaling, dephosphorylation and nuclear translocation of FOXO1, and enhanced recruitment of this tumor suppressor protein to target genes. By performing chromatin immunoprecipitation-sequencing (ChIP seq) analysis, we identified over 800 newly emerged FOXO1-bound genomic loci, including multiple pro-apoptotic BCL2 family proteins (BAX, BAK1, BIK, BBC3, BMF and NOXA1). FOXO1 localization and transcriptional differences were confirmed by ChIP PCR and RT-PCR respectively. Protein levels were measured with Western blotting. BAX was identified as the most common direct target of FOXO1-transriptional activity that was upregulated on both a transcript and protein level. This led us to hypothesize that PRMT5 inhibition could potentially drive a therapeutic vulnerability to the BCL-2 inhibitor venetoclax.

Single agent and combination treatment with venetoclax and PRT382 was performed in nine MCL lines. Of the nine lines, four were considered relatively resistant to PRT-382 and five resistant to venetoclax. Synergy scores, determined from MTS assays, showed significant levels of synergy in the majority of MCL lines tested. CCMCL1 and UPN1, BCL-2 negative MCL lines, and Maver1, which is highly resistant to PRMT5i, were the only cell lines to not show synergy. The cell line with the highest levels of synergy, Z-138, expressed high levels of BCL-2 and is ibrutinib resistant. Overall, there was a strong positive correlation between BCL-2 expression and synergy score (r= -0.8956, p=0.0064).

The synergy seen was confirmed to be through the intrinsic apoptotic pathway based on caspase activity. To determine a mechanism of action, BAX and BAK1 were knocked down in four cell lines, three that displayed synergy and one that was resistant. This suggests that BAX expression is essential for synergy between PRMT5 and BCL2 inhibition to occur. Knock down of BAK1, the other effector of the BCL2 family of proteins, did not show protection suggesting that BAX is necessary and sufficient for this therapeutic synergy to occur. We also determined that p53 status did not correlate to the response seen (p=0.477), supporting that this mechanism is occurring through FOXO1 transcriptional regulation.

In vivo evaluation in two preclinical MCL models showed therapeutic synergy with combination venetoclax/PRT382 treatment. Mice were treated with sub-therapeutic doses of venetoclax and/or PRT382 and disease burden was assessed weekly via flow cytometry. Combination treatment with well-tolerated doses of venetoclax and PRMT5 inhibitors in the MCL in vivo models showed synergistic anti-tumor activity. Both PDX models showed an extension of life with combination treatment (P<0.001) and delayed disease progression (P<0.05). This data provides mechanistic rationale while demonstrating therapeutic synergy in this preclinical study and justifies further consideration of this combination strategy targeting PRMT5 and BCL2 in MCL in the clinical setting.