Synergistic Synthetic Lethal Drug Combinations for E-Cadherin-Deficient Cancers

Abstract

E-cadherin is a cell-to-cell adhesion protein encoded by the gene CDH1. Germline inactivating mutations in CDH1 underpin hereditary diffuse gastric cancer (HDGC), an autosomal dominant disease that strongly increases an individual’s likelihood of developing diffuse gastric cancer (DGC) and invasive lobular breast cancer (ILBC). Current clinical intervention strategies for CDH1 mutation carriers are limited, with the only available approaches being endoscopic surveillance or prophylactic gastrectomy. There is a pressing need to develop chemoprevention strategies to supplement the current approaches.

As E-cadherin is a tumour suppressor protein and lost from the cancer cell, it is not a conventional drug target. This issue can, however, be circumvented using a synthetic lethal approach. Synthetic lethality is defined as a genetic interaction in which a combination of mutations in two or more genes leads to cell death but no death with either mutation alone. Therapeutically, this approach allows for the specific targeting of cells harbouring tumour suppressor gene mutations.

We have previously applied a synthetic lethal strategy to identify candidate therapeutic drugs in a non-malignant isogenic pair of MCF10A cell lines, one with and one without CDH1 expression (MCF10A CDH1-/-). From this testing, a panel of lead drugs were identified, however, the synthetic lethal effects were relatively modest and also reduced the viability of E-cadherin-expressing MCF10A cells. This led to our interest in developing synergistic, synthetic lethal drug combinations, where co-treatment of an additional drug may increase the synthetic lethal response whilst minimising toxicity to wild-type MCF10A cells.

Prior to drug testing, a range of endpoint (CellTiter-Glo, resazurin reduction, and nuclei counting) and real-time (IncuCyte and xCELLigence) cell viability assays were compared to develop an optimum drug screening platform for the MCF10A isogenic cells. We identified nuclei counting and IncuCyte assays as the most cost-effective and accurate approaches. These assay methodologies were used alongside the Chou-Talalay median effect analysis to assess drug synergy at 50% viability loss (ED50) via a combination index (CI), whereby a CI score below 0.9 represents a synergistic interaction.

A range of drugs were delivered simultaneously to the MCF10A isogenic pair in order to discover synergistic combinations that were synthetic lethal with CDH1. The lipid- lowering agents statins were identified as a drug class that specifically reduced MCF10A CDH1-/- cell viability whilst modestly affecting wild-type MCF10A cells up to a concentration of 5 μM. Co-treatment of atorvastatin with our previously identified lead synthetic lethal drugs, entinostat, vorinostat, and saracatinib, maintained synthetic lethality and showed evidence of a synergistic interaction, producing MCF10A CDH1-/- ED50 CI values of 0.37, 0.50, and 0.47, respectively. The most potent synergistic and synthetic lethal effect, however, was produced by atorvastatin in combination with the progesterone receptor and glucocorticoid receptor antagonist mifepristone, with MCF10A and MCF10A CDH1-/- ED50 CI values of 0.49 and 0.08, respectively.

The atorvastatin and mifepristone combination exclusively arrested MCF10A CDH1-/-

cells in the G1 phase of the cell cycle. This effect was shown to be independent of mevalonate and Hippo signalling – key pathways known to be affected by statin treatment. In addition to the non-malignant MCF10A breast cell line, the atorvastatin and mifepristone combination produced a synergistic loss of cell viability in two E-cadherin- deficient breast cancer cell lines, the ILBC-derived line IPH-926 (ED50 CI = 0.69) and the triple-negative breast cancer (TNBC)-derived MDA-MB-231 line (ED50 CI = 0.61). To the best of our knowledge, the combination of a statin and mifepristone has not been previously described in the literature.

In summary, this study has discovered statins as a synthetic lethal drug class in a non- malignant cell line lacking CDH1 expression. Co-treatment of atorvastatin and mifepristone produced an enhanced synthetic lethal response, a synergistic drug relationship that was maintained across malignant breast cell lines devoid of CDH1 expression. These findings may serve as a foundation for the development of novel DGC and ILBC treatment options.