P18.17.A FROM STERILITY TO SELECTIVITY: A C. ELEGANS SYNTHETIC LETHALITY SCREEN REVEALS NON-CONSERVED VULNERABILITIES IN ATRX-DEFICIENT GLIOMA

Abstract BACKGROUND ATRX mutations define a clinically significant subset of gliomas with limited targeted treatment options. We hypothesised that genes inducing synthetic sterility with xnp-1 (ATRX orthologue) in C. elegans could uncover actionable vulnerabilities in ATRX-deficient human cancers. To explore this, we performed a cross-species screen linking developmental genetics to cancer cell viability. MATERIAL AND METHODS Synthetic sterility was quantified in xnp-1 mutant worms following RNAi knockdown. Fertility defects were assessed, and candidate genes were mapped to human orthologues. Corresponding small-molecule inhibitors—TAK-981 (SUMO pathway), UNC-0642 (H3K9 methylation), and G007-LK (tankyrase inhibition)—were evaluated in ATRX-deficient HeLa-LT and SF188 cell lines. Cell viability was measured under single-agent and combination treatments with DNA-damaging agents (doxorubicin, bleomycin). RESULTS RNAi of top hits, including gei-17 and set-25, induced significant fertility defects in xnp-1 mutants. However, pharmacological inhibition of the corresponding human pathways failed to produce selective lethality in ATRX-deficient cells, even when combined with DNA damage. Pathway modulation was confirmed (e.g., SUMO suppression by TAK-981), yet no differential sensitivity emerged. These findings suggest that synthetic sterility in C. elegans may reflect germline-specific stress responses not conserved in cancer biology. CONCLUSION Although the candidate compounds identified through C. elegans synthetic sterility screens did not produce selective cytotoxicity in ATRX-deficient human cancer cells, this outcome provides valuable insight into the translational landscape of cross-species screening. The fertility-based phenotype in xnp-1 mutants likely captures gene interactions relevant to germline development and stress responses, which may not be conserved in cancer biology. By reporting these results, we aim to contribute to a more informed and efficient use of model organisms in neuro-oncology drug discovery.