Abstract B023: Wild-type p53 transcriptional (Tr) readout as a vetting tool for synthetic lethality (SL) screens in cancer therapy

Abstract Background: Some oncogenes are physiologically indispensable (e.g., Myc) or have proven difficult to pharmacologically inhibit (e.g., Ras). SL is an attractive novel approach to drug such oncogenes. The Broad’s DepMap CRISPR screen (CS) suggests SL is NOT monogenic for Ras, so a signature of cooperating genes need to be identified for CS experiments. Genetically engineered mouse models confirmed that mutated (mu) p53 leads to full oncogenic Ras transformation. It appears rational, therefore, that the cellular adjustments needed to accommodate the constant signaling of mu-Ras are provided, at least in one version, by the mu-p53 alterations. Cancers frequently harbor the p53 missense mutation but the loss of function (on Tr targets) is incomplete. We had interrogated the canonical p53 Tr genes in mu-Kras in pancreatic cancer (PC) searching for putative SL candidates. We intentionally included a set of WT-p53 cases co-existing with mu-Kras PC and noticed remarkable overlap suggesting a phenocopy phenomenon. This would elevate the rank of a gene if expressed in mu- and WT-p53. We hypothesized that WT-p53 cancers would phenocopy mu-p53 contributions beyond PC. If confirmed, this method can be applied to strengthen vetting of putative synthetic lethality targets. Methods: We probed the Broad public database for expression of the canonical p53 Tr targets that retained expression. MDM2 represented the most rational internal control for reduced Tr in basal, unstimulated conditions. Retained genes were those with expression level >MDM2. We examined mu-Ras in colon and lung cancers, and Braf V600E in melanoma. Tr readouts seem comparable regardless of p53 mutational status or prevalence. WT-p53 Tr panel is juxtaposed for each oncogene and reveals marked Tr lesion despite the WT sequence of the p53 gene. This may reflect strict needs of an oncogene in a given cell type or the dominant, non-circumventable inhibitory role of p53 network. Results: B-raf V600E melanoma cooperates with mu-p53 less than K-ras but appears to phenocopy the WT-p53 Tr readout. Retained genes are: RPS27L, CDKN1A, Bax, GDF15 & PTP4A1 in melanoma; SERPINB5, Bax, CCNG1, TM7SF3, TRIAP1, RPS27L, PTP4A1, & TRAF4 in colon cancer and TM7SF3, TRAF4, BAX, PTP4A1, CCNG1, ISCU, PLK2, TRIAP1, RPS27L in lung cancer. Notably, WT-p53 cases reveal unexpected vast loss of Tr activity. Few retained genes are restricted to the mu-p53 subset. This method represents a rational filter identify and rank Ts for SL consideration. Some targets are common to more than one malignancy. Conclusions: In lung, colon cancers and melanoma, we find remarkable readout similarities in mu:WT-p53 pairs driven by the same oncogene. This reflects the narrow spectrum of cellular demands of oncogenes such that cancer will need to generate those adjustments in phenocopies. Bax; PTP4A1; RPS27L appear in all screens. This has a major implication narrowing the search for the combinatorial SL targets for CS experiments. Citation Format: Farah Mazahreh, A. Mazin Safar. Wild-type p53 transcriptional (Tr) readout as a vetting tool for synthetic lethality (SL) screens in cancer therapy [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr B023.