TP73 As Novel Determinant of Resistance to BCL-2 Inhibition in Acute Myeloid Leukemia

• Published in: Blood 2019-11, Vol.134 (Supplement_1), p.1251-1251
• Main Authors: Nishida, Yuki, Ishizawa, Jo, Ruvolo, Vivian, Wang, Feng, Takahashi, Koichi, Mak, Po Yee, Carter, Bing Z, Andreeff, Michael
• Format: Article
• Language: English
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2019-127182

Background. BCL-2 inhibition is a novel and highly effective treatment modality in acute myeloid leukemias (AML). AML patients with IDH1/2 mutations are highly sensitive to BCL-2 inhibition by venetoclax (VEN) (Chen et al Nat Med 2015). High expression levels of the BCL-2 family proteins MCL-1 or BCL-XL, or knockout of TP53 have been reported to confer resistance to BCL-2 inhibition (Pan et al. Cancer Cell 2017, Nechiporuk et al. Cancer Discov 2019). p73 is one of the p53 family transcription factors and generates two isoforms, transactivation p73 (TAp73) and the N-terminally truncated ΔNp73. TAp73 shares a homologous N-terminal activation domain with p53 and has pro-apoptotic function similar to p53. ΔNp73 lacks an activation domain and has a dominant negative effect on the DNA binding of TAp73 and more importantly, of p53.TP73 is expressed in AML except in acute promyelocytic leukemias. However, the associations of TP73 isoforms with clinical and genetic characteristics or sensitivity to BCL-2 inhibition in AML have not been explored. Results. We determined copy numbers of TAp73 and ΔNp73 mRNA levels in AML samples (N = 78) and normal CD34+ hematopoietic cells (HPC) using droplet digital PCR and investigated their clinical and biological relevance. Both TP73 isoforms were expressed in AML, with TAp73 expression being 50-fold higher in AML than in CD34+ HPC (P = 0.027); no difference seen for ΔNp73 (P = 0.80), suggesting that TAp73 is aberrantly expressed in AML cells. ΔNp73 and TAp73 mRNA levels were highly correlated (R2 = 0.72, P < 0.0001). AML samples had 10-fold more abundant TAp73 than ΔNp73 mRNA levels (P = 0.0017) and isoforms were not associated with disease status (de novo vs relapsed/refractory) or cytogenetic groups, and were mutation-agnostic, except for IDH1/2. IDH1/2 mutant AML showed lower levels of TAp73 and ΔNp73 than those with wild-type IDH1/2 (P = 0.06 and P = 0.007 for TAp73 and ΔNp73, respectively). In a separate dataset, we observed repressed TP73 in IDH1/2 mutant vs. wild-type AML samples (P = 0.073) by RNAseq analysis (N = 47). Mechanistically, treatment with cell permeable octyl-(R)-2HG, the oncometabolite of mutant IDH1/2, reduced both TAp73 and ΔNp73 and increased susceptibility to VEN. Lentiviral knockdown of p73 in OCI-AML3 cells resulted in enhanced sensitivity to VEN with no significant changes in MCL-1 and p53 protein levels, or TP53 targets (MDM2, CDKN1A, FAS and BBC3). VEN resistant AML cells (MOLM-13 and MV4;11) genera