BackgroundThe tumor suppressor TP53 is the most frequently mutated gene in solid tumors. Although TP53 decides cell fate and governs initiation of apoptosis, inhibitors targeting mutant TP53 did not yet reached clinical use. Our goal was to identify new potential therapeutic targets in TP53 mutant solid tumors by in silico analysis of multiple large, independent next-generation sequencing and gene chip datasets. MethodsFirst, gene expression and mutation data from multiple solid tumors were collected from TCGA and METABRIC databases. Samples were separated based on TP53 mutation status, mutational type and tumor type to identify targetable genes. Differential gene expression was compared using Mann-Whitney test between the mutated (disruptive mutations only) and wild type patient cohorts across all genes. Then, the prognostic value of identified genes was validated in a gene chip-based dataset obtained from the GEO repository. Survival analysis was performed using Cox proportional hazards regression. Significance threshold was set at p<0.01. Finally, False Discovery Rate was computed to correct for multiple hypothesis testing. ResultsThe TCGA dataset include 9,720 patients (21 different cancer types), the Metabric dataset (breast cancer) 1,399 patients, and the GEO dataset (breast, lung, and brain tumors) 7,386 patients. Only genes with higher expression in the TP53 mutant cohort were selected and the list of the top targets was further filtered to include only druggable kinases. The best performing kinases include MPS1 (p=2.9E-58, FC=2.82), PLK1 (p=2.6E-55, FC=2.55), MELK (p=5.2E-54, FC=2.81), and AURKB (p=2E-53, FC=3.23). Each of these kinases had a significant prognostic power as well. Of the top 2 (MPS1 and PLK1), both have multiple inhibitors available (for other indications) with PLK1 closest to the clinical use. ConclusionsOur results suggest that MPS1 (monopolar spindle 1 kinase) and PLK1 (polo like kinase 1) kinases are the strongest druggable targets in TP53 mutant solid tumors. Legal entity responsible for the studySemmelweis University. FundingNational Research, Development and Innovation Office, Hungary. DisclosureAll authors have declared no conflicts of interest.