In our previous work, we performed a comprehensive analysis of DNA sequences of 7,651 tumors from 28 cancer types to identify the transcription factor cut-like homeobox 1 (CUX1) as a novel haploinsufficient tumor suppressor gene (TSG)1, contrary to its assigned oncogenic roles that were known at that time2,3. In our study, we searched for genes showing an enrichment in loss-of-function mutations to provide us with a landscape of putative TSGs, including those occurring at a low frequency such as CUX1, which would have otherwise escaped detection in smaller scale sequencing studies. Specifically, we determined that truncating mutations in one allele of the CUX1 gene were recurrent in myeloid malignancies such as AML1. Notably, CUX1 lies in a region of chromosome 7q, which is recurrently lost through complete or partial monoallelic deletions [-7/del(7q)] in several types of myeloid malignancies, implying that CUX1 is an important chromosome 7q target gene in these diseases4.
Following the identification of CUX1 as a haploinsufficient TSG, we were faced with two important questions that we sought to address in the present study: firstly, was it possible to identify vulnerabilities specific to CUX1-deficient AML that could be exploited for AML therapy? Since both -7/del(7q) lesions and CUX1 haploinsufficiency confer a poor prognosis in AML and other myeloid malignancies1,5, there is an unmet clinical need for targeted therapies in these high-risk cases. Secondly, how did CUX1 haploinsufficiency contribute to leukemia development and did it require cooperating mutations? While the role of biallelic TSG inactivation in tumorigenesis has been widely acknowledged since Knudson`s two-hit hypothesis from 19716, the function of TSG haploinsufficiency in cancer remains less well described, partly due to the lack of genetic mouse models, which can recapitulate haploinsufficiency accurately in vivo7,8.
To address the first question, we performed CRISPR/Cas9 drop-out screens in isogenic wild-type and CUX1-knockout AML cell lines, which identified the CASP8 and FADD-like apoptosis regulator (CFLAR) gene as a preferential vulnerability in CUX1-deficient cells (Figure 1). The encoded anti-apoptotic protein CFLAR (also known as cFLIP) inhibits the extrinsic apoptosis pathway by competing with pro-caspase-8 for binding to death receptors, thereby preventing caspase activation9. Mechanistically, we could show that the full-length isoform of CUX1 acts as a direct transcriptional repressor at the CFLAR promoter, providing insight into how CUX1 deficiency upregulates CFLAR expression to safeguard against apoptosis.
To address the second question, we next sought to generate a conditional Cux1-haploinsufficient mouse model. However, we quickly realized that heterozygous Cux1 deletion alone was insufficient to promote AML development, suggesting the requirement for additional genetic mutations. Therefore, we searched for mutations cooccurring with -7/del(7q) lesions in human AML and chose FLT3-internal tandem duplication (FLT3ITD) mutations, which are found in ~10% of -7/del(7q) cases of AML10, for further investigation. Combining Cux1 haploinsufficiency with Flt3ITD mutation resulted in a lethal leukemia in all double-mutant mice (Cux1+/-;Flt3ITD) with a median survival of 28 weeks. Interestingly, while moribund Cux1+/-;Flt3ITD mice culled at an earlier age were diagnosed with a disease similar to human chronic myelomonocytic leukemia (CMML), older animals exhibited features of AML (Figure 2). In line with our previous CRISPR/Cas9 screen in human cells, genetic depletion of Cflar preferentially diminished the survival of Cux1-deficient cells in vitro and in vivo, supporting our hypothesis that Cflar is a selective vulnerability in the context of Cux1 deficiency. Combined, these results indicate that CUX1 haploinsufficiency cooperates with FLT3ITD mutation to promote AML, which is associated with an induction in CFLAR-mediated anti-apoptosis pathways.
To investigate whether targeting of CFLAR can be translated into AML therapy, we wanted to assess the effects of pharmacological CFLAR inhibition on CUX1-deficient cells. However, since direct CFLAR inhibitors are not routinely available, we tested the impact of the second mitochondrial-derived activator of caspases (SMAC)-mimetic drug birinapant (https://pubchem.ncbi.nlm.nih.gov/compound/Birinapant) on Cux1-haploinsufficient cell survival as an alternative therapeutic approach (Figure 3). Indeed, murine hematopoietic cells from Cux1-haploinsufficient mice as well as primary human AML samples harboring -7/del(7q) lesions displayed higher sensitivity to birinapant compared with their wild-type and normal karyotype counterparts, respectively.
Finally, considering that birinapant in combination with azacitidine for the treatment of genetically unselected patients with higher-risk myelodysplastic syndrome or CMML failed to improve response rates compared with azacitidine alone12, our findings suggest that clinical studies evaluating SMAC-mimetics such as birinapant could benefit from a more stringent stratification of patients. Therefore, we encourage not only the development of direct CFLAR inhibitors but also the further clinical evaluation of SMAC-mimetics specifically in CUX1-haploinsufficient and -7/del(7q) myeloid malignancies.
- Wong, C. C. et al. Inactivating CUX1 mutations promote tumorigenesis. Nat Genet 46, 33-38, doi:10.1038/ng.2846 (2014).
- Cancer Genome Atlas, N. Comprehensive molecular characterization of human colon and rectal cancer. Nature 487, 330-337, doi:10.1038/nature11252 (2012).
- Michl, P. et al. CUTL1 is a target of TGF(beta) signaling that enhances cancer cell motility and invasiveness. Cancer Cell 7, 521-532, doi:10.1016/j.ccr.2005.05.018 (2005).
- Jerez, A. et al. Loss of heterozygosity in 7q myeloid disorders: clinical associations and genomic pathogenesis. Blood 119, 6109-6117, doi:10.1182/blood-2011-12-397620 (2012).
- Grimwade, D. et al. Refinement of cytogenetic classification in acute myeloid leukemia: determination of prognostic significance of rare recurring chromosomal abnormalities among 5876 younger adult patients treated in the United Kingdom Medical Research Council trials. Blood 116, 354-365, doi:10.1182/blood-2009-11-254441 (2010).
- Knudson, A. G., Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A 68, 820-823, doi:10.1073/pnas.68.4.820 (1971).
- Venkatachalam, S. et al. Retention of wild-type p53 in tumors from p53 heterozygous mice: reduction of p53 dosage can promote cancer formation. EMBO J 17, 4657-4667, doi:10.1093/emboj/17.16.4657 (1998).
- Berger, A. H. & Pandolfi, P. P. Haplo-insufficiency: a driving force in cancer. J Pathol 223, 137-146, doi:10.1002/path.2800 (2011).
- Irmler, M. et al. Inhibition of death receptor signals by cellular FLIP. Nature 388, 190-195, doi:10.1038/40657 (1997).
- McNerney, M. E. et al. The spectrum of somatic mutations in high-risk acute myeloid leukaemia with -7/del(7q). Br J Haematol 166, 550-556, doi:10.1111/bjh.12964 (2014).
- Cassier, P. A., Castets, M., Belhabri, A. & Vey, N. Targeting apoptosis in acute myeloid leukaemia. Br J Cancer 117, 1089-1098, doi:10.1038/bjc.2017.281 (2017).
- Donnellan, W. B. et al. A phase 2 study of azacitidine (5-AZA) with or without birinapant in subjects with higher risk myelodysplastic syndrome (MDS) or chronic myelomonocytic leukemia (CMML). Journal of Clinical Oncology 34, doi:10.1200/JCO.2016.34.15_suppl.7060 (2016).