Therapeutic developments targeting acute myeloid leukemia (AML) have been in the pipeline for five decades and have recently resulted in the approval of multiple targeted therapies. However, there remains an unmet need for molecular treatments that can deliver long-term remissions and cure for this heterogeneous disease. Previously, a wide range of small molecule drugs were developed to target sub-types of AML, mainly in the relapsed and refractory setting; however, drug resistance has derailed the long-term efficacy of these as monotherapies. Recently, the small molecule venetoclax was introduced in combination with azacitidine, which has improved the response rates and the overall survival in older adults with AML compared to those of chemotherapy. However, this regimen is still limited by cytotoxicity and is not curative. Therefore, there is high demand for therapies that target specific abnormalities in AML while sparing normal cells and eliminating leukemia-initiating cells. Despite this, the urgent need to develop these therapies has been hampered by the complexities of this heterogeneous disease, spurring the development of innovative therapies that target different mechanisms of leukemogenesis. This review comprehensively addresses the development of novel targeted therapies and the translational perspective for acute myeloid leukemia, including the development of selective and non-selective drugs.

Background: XPO1 is a nuclear export receptor responsible for exporting >200 proteins out of the nucleus, including many cancer-related proteins. We previously reported recurrent hotspot mutations occurring in XPO1 in cancer and identified XPO1 R749Q variant only in solid tumors. To determine the prevalence of XPO1 mutations across cancer types, we performed a large-scale genomic analysis of 217,570 patients with cancer to identify and characterize XPO1 variants from real-world patient tumors. Methods: Solid tumor samples representing 14 cancer types were submitted to Caris Life Sciences (Phoenix, AZ) for Next-generation sequencing of DNA (592 genes or WES). TMB-High (defined as >10 mt/MB) and dMMR/MSI-H was tested by IHC/NGS. We engineered CRISPR knock-in HCT116 and LS-174T colon cancer isogenic cell lines bearing XPO1 R749Q mutations and Stochastic Optical Reconstruction Microscopy (STORM) analysis was performed. Structural modeling was performed with published XPO1 structures by mapping XPO1 mutations from the MSKCC-IMPACT dataset. Results: A total of 96 patients with XPO1 R749Q mutations were identified (50% occurred in endometrial cancer and 23% occurred in colorectal cancer). TMB-H was observed in 92% of XPO1 R749Q mt tumors (9% were dMMR/MSI-H), while 86% of XPO1 R749Q mt tumors were POLE-mutated within the endonuclease domain. Median variant allele frequency for XPO1 R749Q mt and POLE mt were 25% and 29%, respectively. STORM imaging revealed that XPO1 R749Q mutant cells had significant localization of XPO1 in the cytoplasm compared to XPO1 WT cells, especially at the outside edge of the nuclear pores. Structural modeling predicted that XPO1 R749Q mt affected the regulatory H9-loop of XPO1 favoring increased shuttling and retention in the cytoplasm. Mass spectrometry analysis of nuclear and cytoplasmic fractionated proteomes confirmed that XPO1 R749Q mt cells had increased export of proteins compared to XPO1 WT cells. A library screen of >200 FDA-approved drugs revealed a strong therapeutic resistance of XPO1 R749Q cells, especially to chemotherapies used in the treatment of colon cancer. XPO1 inhibition with selinexor synergized with chemotherapy in XPO1 R749Q mt cells in vitro and overcame resistance to irinotecan in vivo in xenograft mice models. Conclusion: This study sheds novel insights into the role of nuclear export in cancers. Specifically, XPO1 R749Q mutations are enriched in TMB-H endometrial and colon cancers and increase nuclear export of key proteins that confer resistance to therapies using DNA-damaging agents. Our current work aims to identify novel therapies that can overcome resistance to DNA-damaging therapies seen in XPO1 mutant cells, such as immune checkpoint inhibitor (ICI) therapy given the co-occurrence with POLE mutations and recent reports of high level of response to ICI therapy in TMB-H and POLE mt colorectal cancers. Citation Format: Tulasigeri M. Totiger, Yasmine Baca, Wannasiri Chiraphapphaiboon, Sana Chaudhry, Skye Montoya, Monika Chojnacka, Gabriel Gaidosh, Jumana Afaghani, Maurizio Affer, Christopher D. Armstrong, Ryan Notti, Jenna Zabroski, Jacob Jahn, Vindhya Nawaratne, Ramiro Verdun, Hai Dang Nguyen, Chadi Nabhan, Thomas J. Herzog, Phil Walker, Andrew Elliott, Emil Lou, Wafik S. El-Deiry, Edith Mitchell, Jose Antonio Rodriguez, Justin Taylor. Pan-cancer analysis of XPO1 R749Q mutations across 217,570 patients reveals association with high tumor mutational burden and therapy resistance. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4253.

XPO1 is a nuclear export receptor responsible for exporting many key proteins critical for cell survival from the nucleus to the cytoplasm. Recently, we identified a highly statistically significant hotspot mutation in XPO1 R749Q occurring in patients with colorectal cancer. Out of 96 patients identified to have XPO1 R749Q mutations, 27 (28%) were in colorectal cancer (CRC). 95% of XPO1 R749Q mutant CRC were tumor mutation burden-high (TMB-H; >10 mut/Mb) and 0 were mis-match repair deficient (dMMR)/micro satellite instability-high (MSI-H). Co-mutation analysis showed that 95% of XPO1 R749Q CRC were POLE mutated, which was significantly enriched over XPO1 WT tumors. To investigate whether XPO1 R749Q might play a role in CRC biology, we have generated two different isogenic colorectal cancer cell lines bearing the XPO1 R749Q mutation (HCT116 and LS174T) using CRISPR-CAS9 to better understand the role of this aberrant XPO1 mutation on nuclear export and tumorigenesis. Immunofluorescence (IF) and Stochastic Optical Reconstruction Microscopy (STORM) super-resolution experiments showed that R749Q mutant cells had significantly increased localization of XPO1 in the cytoplasm of the XPO1 R749Q mutant cells when compared to wildtype (XPO1 WT) cells, particularly at the edge of the nuclear pores of the XPO1 R749Q mutant cells suggesting increased nuclear export. Mass spectrometry analysis of nuclear and cytoplasmic fractionated proteins confirmed that XPO1 R749Q mutant cells had increased export of proteins from the nucleus compared to XPO1 WT cells. Structural modeling using published structures of XPO1 bound to Ran-GTP predicted that XPO1 R749Q mutation increased affinity of the regulatory H9-loop of XPO1 to Ran, thus favoring increased shuttling and retention in the cytoplasmic compartment. Therefore, we hypothesize that XPO1 R749Q may represent a novel mechanism of nuclear export alteration involved with kinetic tuning of global transport. Despite the increased nuclear export function, XPO1 R749Q cells did not show a proliferative advantage. We thus set out to test whether XPO1 R749Q might be enriched in cancer as an adaptive response to chemotherapy treatment. Using a chemical compound library of >200 FDA-approved cancer therapies we observed a strong therapeutic resistance of XPO1 R749Q cells relative to XPO1 WT cells, specifically to chemotherapies used in the treatment of colon cancer, such as irinotecan. However, these cells remained sensitive to treatment with the XPO1 inhibitor selinexor. Mice xenografted with XPO1 R749Q mutant HCT116 cells showed moderate tumor response to selinexor or irinotecan monotherapies but prolonged tumor responses to combination therapy. Recent clinical data has shown POLE mutations to be a biomarker for response to PD-1 immune checkpoint blockade, especially in CRC. Since our genetic data identify POLE mutations commonly co-occurring with XPO1 R749Q in colon cancer, we are planning to test the combination of PD-1 inhibitors and selinexor to target both these mutations, as well as CRC tumors without these mutations. Citation Format: Tulasigeri M. Totiger, Sana Chaudhry, Skye Montoya, Monika Chojnacka, Gabriel Gaidosh, Jumana Afaghani, Maurizio Affer, Jenna Zabroski, Jacob Jahn, Vindhya Nawaratne, Ryan Notti, Ramiro Verdun, Daniel Bilbao, Josean Rodriguez, Justin Taylor. Novel therapeutics for targeting the aberrant nuclear export machinery in colorectal cancer [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr B018.