Targeting DNA repair vulnerabilities to eradicate TP53 mutated AML.

Abstract Mutations in TP53 (TP53mut) confer an adverse prognosis in AML. Current treatment strategies, including conventional chemotherapy, hypomethylating agents, and venetoclax-based therapies, have shown limited efficacy in TP53mut AML, with low response rates and poor overall survival. Allogeneic hematopoietic stem cell transplantation is a potentially curative option; however, its efficacy in TP53mut AML depends on comorbid conditions and disease status at transplantation. Novel therapeutic modalities, including immune-based therapies, did show promise in early-phase studies but did not yet translate into effective therapies in randomized controlled trials. These reports demonstrate a significant need for improved treatment for this difficult-to-treat population. AML cells carrying TP53-/- and 6 clinically relevant TP53muts (R175H, Y220C, M237I, R248Q, R273H, andR282W) accumulated spontaneous DNA damage, including highly lethal DNA double strand breaks (DSBs). TP53 is active at the hub of the DNA damage response (DDR). An increasing number of genes regulated by TP53 transcriptional functions have been shown to directly or indirectly affect the DDR. To identify DDR mechanisms responsible for DNA repair and survival of TP53mut AML cells, three unbiased approaches have been applied: (1) Crispr/Cas9 screen using the gRNA library of 365 DDR genes; (2) RNAseq dataset analysis of the expression of 1,800 DDR-related genes; and (3) drug sensitivity screen to fifteen DDR inhibitors. One of the top hits from Crispr/Cas9 screen was BRCA2 which promotes nuclear localization of RAD51 to stimulate and maintain strand invasion, a vital step of homologous recombination repair (HRR). Remarkably, RAD51 protein is strongly overexpressed and HRR is stimulated in AML clones carrying TP53-/-and all 6 TP53muts consistent with transcriptional repression of RAD51 by TP53. Therefore, HRR may play a pivotal role in protecting TP53mut AML cells from abundant highly toxic spontaneous DSBs. In concordance, drug sensitivity screen revealed that AML TP53-/- and 6 TP53muts clones were hypersensitive to RAD51inhibitor (RAD51i) when compared to TP53 wild-type AML cells and normal bone marrow cells. RNAseq database analysis validated by Western blot revealed downregulation of XPC and DDB2 and inhibition of the nucleotide excision repair (NER) in TP53-/- and all six TP53muts AML cell lines consistent with lack of TP53-mediated transcriptional stimulation. XPC and DDB2 are the key proteins in the global genome NER (GG-NER). GG-NER removes bulky melphalan (a nitrogen mustard used in the management of multiple myeloma) DNA monoadducts, but not irofulven-mediated DNA damage which is recognized by transcription-coupled NER (TC-NER). In concordance, TP53-/- and TP53mut AML cells were selectively sensitive to melphalan, but not to irofulven. Finally, the combination of melphalan and RAD51i exerted a tremendous inhibitory effect against AML cell lines and primary AML patient cells carrying TP53-/- and TP53muts while sparing most of the normal hematopoietic cells. We are currently testing the combination of melphalan and RAD51i in mice carrying TP53mut AML. We pinpointed the “Achilles heels” of DDR in TP53mut AMLs: stimulation of HRR and downregulation of GG-NER. These weaknesses were successfully targeted by the combination of RAD51i which attacks HRR and melphalan which explores GG-NER deficiency.