SKI complex loss renders 9p21.3-deleted or MSI-H cancers dependent on PELO

Abstract

Cancer genome alterations often lead to vulnerabilities that can be used to selectively target cancer cells. Various inhibitors of such synthetic lethal targets have been approved by the FDA or are in clinical trials, highlighting the potential of this approach1–3. Here we analysed large-scale CRISPR knockout screening data from the Cancer Dependency Map and identified a new synthetic lethal target, PELO, for two independent molecular subtypes of cancer: biallelic deletion of chromosomal region 9p21.3 or microsatellite instability-high (MSI-H). In 9p21.3-deleted cancers, PELO dependency emerges from biallelic deletion of the 9p21.3 gene FOCAD, a stabilizer of the superkiller complex (SKIc). In MSI-H cancers, PELO is required owing to MSI-H-associated mutations in TTC37 (also known as SKIC3), a critical component of the SKIc. We show that both cancer subtypes converge to destabilize the SKIc, which extracts mRNA from stalled ribosomes. In SKIc-deficient cells, PELO depletion induces the unfolded protein response, a stress response to accumulation of misfolded or unfolded nascent polypeptides. Together, our findings indicate PELO as a promising therapeutic target for a large patient population with cancers characterized as MSI-H with deleterious TTC37 mutations or with biallelic 9p21.3 deletions involving FOCAD. Analysis of large-scale CRISPR screening data, combined with experiments in patient-derived tumour organoid models, identifies PELO as a potential therapeutic target in chromosomal 9p21.3-deleted cancers and microsatellite-unstable cancers harbouring specific mutations.