XPO1 is a nuclear export receptor belonging to the importin-β superfamily of karyopherins and plays a central role in mediating the nuclear export of cargo proteins by recognizing their NES (13,14). Notably, XPO1 is the only known exportin responsible for transporting leucine-rich NES-containing proteins from the nucleus to the cytoplasm, a function that is critically dependent on its distinct structural architecture. XPO1 harbors a specialized binding pocket that specifically recognizes and engages NES motifs (7). These motifs are typically short, hydrophobic peptide sequences enriched in leucine residues and the precise amino acid composition and three-dimensional conformation of the NES-binding domain allow XPO1 to form complementary interactions with the hydrophobic side chains within the NES, akin to a lock-and-key mechanism, enabling selective binding to NES-bearing proteins or ribonucleoprotein (RNP) complexes (15,16). In addition, XPO1 contains a high-affinity binding site for Ran-GTP, a small GTPase that regulates nucleocytoplasmic transport (17). The association of Ran-GTP with XPO1 induces a conformational change that stabilizes the interaction between XPO1 and its cargo, facilitating the formation of the export-competent XPO1-cargo-Ran-GTP ternary complex and enabling its translocation through the NPC (18). Upon reaching the cytoplasm, hydrolysis of Ran-GTP to Ran-GDP triggers a structural rearrangement that promotes cargo release, thereby completing the export cycle (19). Furthermore, XPO1 possesses a distinct NPC-interacting domain that mediates transient interactions with phenylalanine-glycine (FG)-repeat nucleoporins located within the NPC. This interaction allows the export complex to dock at and traverse the nuclear pore, ensuring directional transport from the nucleus to the cytoplasm (18).

In summary, based on the aforementioned domains, XPO1 can mediate the nuclear export of RNA or proteins, thereby participating in the regulation of numerous biological processes.

Eukaryotic cells establish functional compartmentalization between the nucleus and cytoplasm through the NPC, which also serves as the primary conduit for nucleocytoplasmic exchange of macromolecules and signaling molecules. Consequently, the NPC plays a key role in maintaining cellular homeostasis and functions (8,15). The NPC is a large, evolutionarily conserved protein complex embedded within the nuclear envelope, exhibiting an octagonal symmetry and a characteristic three-dimensional architecture that resembles a ‘basketball hoop’ when viewed laterally (20). It is primarily composed of multiple copies of ~30 distinct nucleoporins (Nups), which are precisely organized into subcomplexes to form a central transport channel that facilitates selective macromolecular trafficking (21). This central channel permits passive diffusion of small molecules, such as ions and low-molecular-weight metabolites, while the translocation of larger macromolecules, including proteins and RNA species, requires active, signal-dependent transport mechanisms mediated by nuclear transport receptors (17).

The XPO1-mediated nuclear export process proceeds through a well-orchestrated sequence of molecular events. First, substrate recognition occurs: XPO1 specifically identifies cargo proteins containing NES (22). The NES is a short peptide motif enriched in hydrophobic amino acids, particularly leucine, and typically spans 10–30 residues (10). XPO1 binds tightly to the NES via its dedicated binding domain, initiating the export process. This interaction is highly selective, ensuring that only cargoes bearing functional NES motifs, such as specific proteins or RNP complexes, are recruited for nuclear export (23). Second, formation of the export complex takes place: Upon substrate binding, XPO1 associates with Ran-GTP, a small GTPase that acts as a key regulatory cofactor, to assemble a stable trimeric export complex (XPO1-cargo-Ran-GTP) (24). The binding of Ran-GTP induces a conformational change in XPO1 that enhances its affinity for the cargo and stabilizes the complex, thereby rendering it translocation-competent (25). Third, docking and translocation through the NPC occur: The assembled trimeric complex is directed to the NPC, where XPO1 interacts with phenylalanine-glycine (FG)-repeat nucleoporins via its NPC-binding domain, facilitating docking at the nuclear side of the pore. The NPC, a large transmembrane channel embedded in the nuclear envelope, serves as the sole gateway for nucleocytoplasmic transport. Through transient interactions with FG-Nups, the complex undergoes facilitated translocation across the NPC toward the cytoplasm (22). Finally, cargo release and recycling are achieved in a GTP hydrolysis-dependent manner. Upon reaching the cytoplasm, Ran-GTP is hydrolyzed to Ran-GDP by RanGAP and its cofactors, triggering a conformational rearrangement in XPO1 that reduces its affinity for both the cargo and Ran-GDP. This leads to the dissociation of the cargo into the cytoplasm. Subsequently, XPO1 and Ran-GDP are recycled back to the nucleus for subsequent rounds of export, completing the transport cycle (18,26) (Fig. 1).

The XPO1-mediated nuclear export process plays a pivotal role in cell cycle regulation. For example, cyclin-dependent kinase inhibitors p21 and p27 are actively exported from the nucleus by XPO1, thereby modulating their nuclear accumulation and subsequently regulating CDK activity and cell cycle progression (27,28). Similarly, transcription factors of the FOXO family are constitutively shuttled out of the nucleus by XPO1 under basal conditions, limiting their access to target genes involved in apoptosis and antioxidant defense; upon cellular stress, inhibition of XPO1 leads to nuclear retention of FOXO proteins, enabling them to activate transcriptional programs essential for stress resistance and cell fate control (29,30). The tumor suppressor p53 is also subject to XPO1-dependent regulation, following completion of DNA damage repair, XPO1 mediates the nuclear export of p53 to attenuate its transcriptional activity and prevent prolonged cell cycle arrest or aberrant apoptosis (31,32). Through this dynamic spatial control of key regulatory proteins, XPO1 fine-tunes key cellular processes including proliferation, differentiation and homeostasis. Beyond these functions, XPO1 contributes notably to the regulation of cellular stress responses and inflammatory signaling. In the NF-κB pathway, degradation of IκBα releases NF-κB for nuclear translocation and activation of pro-inflammatory genes, whereas XPO1 counteracts this activation by exporting NF-κB, often as a complex with newly synthesized IκBα, back to the cytoplasm, thus providing negative feedback to limit the duration and intensity of NF-κB signaling and prevent chronic inflammation (33,34). Furthermore, under oxidative stress or heat shock, XPO1 regulates the nucleocytoplasmic trafficking of stress-responsive transcription factors such as HSF1, ensuring timely induction of cytoprotective gene expression programs (35). In addition, XPO1 facilitates the nuclear export of specific mRNAs, particularly those encoding growth regulators and oncoproteins, thereby extending its regulatory influence to post-transcriptional control of gene expression (36,37).

In summary, XPO1 is not only a central mediator of nucleocytoplasmic transport but also a key regulatory hub that integrates intracellular and extracellular signals with essential cellular processes. Maintenance of its normal physiological function is vital for maintaining cellular homeostasis, ensuring proper signal transduction and enabling spatiotemporal regulation of key molecular events. Dysregulation of XPO1 function can result in the mislocalization of key regulatory proteins, thereby disrupting cell fate decisions and contributing to tumorigenesis and cancer progression. Therefore, a comprehensive understanding of the molecular mechanisms underlying XPO1 activity not only advances fundamental knowledge in cell biology but also provides a solid theoretical foundation for developing targeted therapies against its pathological dysregulation.

In various hematological malignancies (such as MM (12), DLBCL (38) or acute myeloid leukemia (39)) and solid tumors (including ovarian cancer (40), gastric cancer (41), pancreatic cancer (42) or glioblastoma (43)), the expression level of XPO1 is markedly higher than that in the corresponding normal tissues. Moreover, multiple clinical studies have confirmed that its high expression is associated with poor prognosis, accelerated disease progression, and chemotherapy resistance in patients (44). A large number of immunohistochemical, RT-qPCR and gene expression profiling data show that the protein and mRNA levels of XPO1 are abnormally upregulated in tumor tissues, suggesting that it is not only a functional driver in the process of tumorigenesis but may also serve as a potential prognostic biomarker (45). Studies have found that XPO1 promotes tumor occurrence and development at multiple levels by mediating the abnormal nuclear export of key TSPs and regulatory factors, thereby disrupting their normal subcellular localization and functional activity (10,26,46,47).

In the p53 signaling pathway, XPO1 actively exports both p53 and its upstream DNA damage sensor kinase ATM from the nucleus, resulting in their aberrant cytoplasmic accumulation and functional inactivation (47,48). Excessive nuclear export of these proteins severely impairs their transcriptional regulatory functions, compromising genomic surveillance, increasing mutational burden and thereby facilitating malignant transformation. This mechanism is particularly important and clinically validated in hematologic malignancies and specific solid tumors harboring wild-type p53, providing a mechanistic explanation for the observed loss-of-function phenotype despite an intact p53 gene (32,49).

XPO1 drives cell cycle dysregulation by mediating the nuclear exclusion of key tumor suppressors. For example, the CDK inhibitors p21 and p27, key negative regulators of the G₁/S transition, exert their inhibitory effects on cyclin-CDK complexes when localized in the nucleus (9); similarly, the retinoblastoma (Rb) protein suppresses S-phase gene expression by sequestering E2F transcription factors (48). Under conditions of XPO1 overexpression, these proteins are constitutively exported from the nucleus, leading to unchecked cell cycle progression and uncontrolled tumor cell proliferation. Moreover, p73 is subject to XPO1-mediated nuclear export and inactivation, further undermining the integrity of the tumor suppressor network (49).

XPO1 suppresses apoptosis by exporting pro-apoptotic factors. The FOXO family of transcription factors (for example, FOXO1 and FOXO3a) induces expression of pro-apoptotic genes under stress conditions; however, elevated XPO1 activity promotes rapid nuclear export and subsequent degradation of FOXO proteins, abolishing their apoptotic function (50,51). Similarly, Bim, a pivotal pro-apoptotic member of the Bcl-2 family, exerts its mitochondrial-stabilizing and caspase-activating effects primarily when retained in the nucleus, but its efficacy is diminished by XPO1-dependent nuclear export (52,53). Notably, survivin can also be exported by XPO1 in specific splice variants, modulating its stability and distribution, thereby indirectly tipping the balance toward cell survival (54).

In inflammatory and proliferative signaling pathways, XPO1 enhances NF-κB activation by exporting IκBα, the endogenous inhibitor of NF-κB, thereby releasing NF-κB from cytoplasmic sequestration and enabling its nuclear translocation and transcriptional activation of pro-inflammatory, anti-apoptotic and pro-migratory genes (50,51,55). Concurrently, XPO1 exports eIF4E from the nucleus, facilitating its cytoplasmic assembly into translation initiation complexes and selectively enhancing the translation of oncogenic mRNAs, including c-Myc, cyclin D1 and VEGF (56–58). In lymphomas, XPO1 overexpression mediates the nuclear export of STAT6, leading to increased levels of phosphorylated STAT6? and accelerated tumor cell proliferation (59). Furthermore, while nuclear accumulation of MYC drives malignant proliferation, XPO1 contributes to MYC oncogenicity by regulating its nucleocytoplasmic shuttling to maintain MYC protein stability (56) (Fig. 2).

In summary, XPO1 drives tumorigenesis through a coordinated and multifaceted mechanism by systematically altering the subcellular localization of key regulatory proteins, thereby disrupting key cellular processes such as the DNA damage response, cell cycle control, apoptosis and signal transduction. This ‘multi-target hijacking’ strategy enables XPO1 to simultaneously impair multiple tumor-suppressive pathways, positioning it as a highly compelling therapeutic target in oncology. Targeting XPO1 therefore offers a dual advantage: it not only restores the functional integrity of diverse tumor suppressor networks but also counters chemotherapy resistance and potentiates the effectiveness of current treatment modalities. As such, XPO1 inhibition represents a strategy with substantial potential for both advancing fundamental cancer biology and enabling meaningful clinical translation.

The SINE selinexor, a representative compound in its class, has demonstrated notable antitumor activity across various hematological malignancies, with its clinical translation achieving notable breakthroughs (67). Particularly in the treatment of refractory or relapsed diseases, selinexor offers a novel therapeutic option for patients who have failed conventional therapies by restoring the nuclear localization and transcriptional function of key TSPs.

The BOSTON Trial (Phase III RCT): Further validation for MM came from the BOSTON trial, a randomized, open-label, phase III study comparing the triplet regimen of once-weekly selinexor, bortezomib and dexamethasone (SVd) vs. the standard doublet Vd (bortezomib plus dexamethasone) in patients with RRMM. The SVd arm demonstrated a statistically notable improvement in median progression-free survival (PFS), with particularly pronounced benefits observed in patients who had received only 1–3 prior lines of therapy. Moreover, the SVd regimen yielded higher overall response rates and complete response rates, indicating synergistic antitumor activity. As a result, the SVd regimen was incorporated into international treatment guidelines as a standard-of-care option for eligible patients with RRMM (64,68,69). A subsequent review further validated these findings by systematically demonstrating that selinexor-containing regimens maintain consistent efficacy across multiple difficult-to-treat MM subpopulations, including triple-class refractory disease, renal impairment, high-risk cytogenetics and prior anti-CD38 therapy. An ongoing phase Ib/II trial is evaluating the quadruplet of selinexor with carfilzomib, isatuximab and dexamethasone in RRMM, while another phase II trial is assessing SVd-like induction specifically in newly diagnosed MM with extramedullary disease. Notably, selinexor is now being explored as maintenance therapy following CAR-T cell treatment, reflecting its expanding role across the MM treatment continuum. The SENTRY Trial (XPORT-MF-034; Phase III RCT): In myelofibrosis, a disease setting previously unexplored for XPO1 inhibition, the randomized, double-blind, placebo-controlled SENTRY trial evaluated selinexor (60 mg once weekly) in combination with ruxolitinib vs. placebo plus ruxolitinib in JAK inhibitor-naïve patients. Results released in March 2026 demonstrated that the trial met its first co-primary endpoint of spleen volume reduction ≥35% (SVR35) at week 24: 50 vs. 28% (one-sided P<0.0001), with rapid, deep and durable spleen responses and a promising overall survival signal. Although the second co-primary endpoint of symptom improvement was not reached, this landmark study positions selinexor as a potential first-in-class targeted therapy in myeloproliferative neoplasms (70).

The STORM Trial (Phase IIb, Single-Arm): In the field of MM, the STORM study enrolled patients with penta-refractory disease, defined as resistance to at least five major drug classes: Proteasome inhibitors, immunomodulatory agents, anti-CD38 monoclonal antibodies, alkylating agents and glucocorticoids. Despite this extensive prior treatment and refractoriness, the combination of selinexor and low-dose dexamethasone achieved an overall response rate (ORR) of ~26%, with some patients attaining a very good partial response or better. Based on these findings, the FDA granted accelerated approval in 2019 for selinexor in combination with dexamethasone for the treatment of RRMM (64,71). The SADAL Trial (Phase II, Single-Arm): In DLBCL, the SADAL study evaluated selinexor monotherapy in patients with relapsed/refractory disease. The trial reported an ORR of 28% in the overall population, with a complete response rate of 12%; notably, patients with the germinal center B-cell-like subtype showed a trend toward improved outcomes. Given this clinical benefit, the FDA approved selinexor in 2020 for the treatment of relapsed/refractory DLBCL in patients ineligible for intensive chemotherapy or CAR-T cell therapy (71–75).

In AML, early-phase I/II studies have demonstrated that selinexor combined with azacitidine or low-dose cytarabine can induce complete remissions in elderly or unfit patients, with particular promise in those harboring TP53 mutations or complex karyotypes. In T-cell lymphomas, including peripheral T-cell lymphoma and mycosis fungoides, small studies have documented stable disease or partial responses (76–83).

In conclusion, selinexor has demonstrated substantial clinical value across a range of hematological malignancies, particularly in patients with highly refractory disease who have limited therapeutic options. Its ability to induce responses in such heavily pretreated populations underscores its distinctive mechanism of action and ongoing therapeutic relevance. As understanding of the drug's pharmacodynamics and predictive biomarkers continues to evolve, there is strong potential for refining patient selection and developing individualized treatment strategies that maximize clinical benefit.

Although research on SINE has advanced more rapidly in hematological malignancies, their clinical exploration in solid tumors has increasingly demonstrated potential antitumor activity and therapeutic value. Selinexor, the first-in-class SINE, is undergoing extensive evaluation in multiple refractory solid tumor types, leveraging its favorable oral bioavailability and robust tissue penetration properties. Although clinical evaluation in solid tumors is ongoing, these indications face more prominent barriers. The clinical efficacy of XPO1 inhibitors in solid tumors is currently less established than in hematological malignancies, frequently characterized by modest PFS gains, limited OS extensions and heightened toxicity profiles (84).

The SEAL Trial (Phase III RCT): In advanced dedifferentiated liposarcoma (DDLS), the SEAL study evaluated the efficacy and safety of selinexor monotherapy in patients who had progressed after multiple prior therapies. Results demonstrated that Selinexor notably prolonged median PFS compared with placebo. However, overall OS did not show a statistically significant difference, a notable clinical limitation given the relentless course characteristic of DDLS. Mechanistically, DDLS is frequently driven by amplification of the 12q13-15 region, resulting in overexpression of MDM2 and CDK4; inhibition of XPO1 prevents the nuclear export and subsequent degradation of TSPs, including p53 and Rb, thereby partially restoring their activity (54,85).

In gynecologic malignancies, the phase III SIENDO trial evaluated oral selinexor as maintenance therapy after first-line chemotherapy for advanced or recurrent endometrial cancer. The primary analysis showed a statistically significant improvement in median PFS (HR, 0.71; P=0.049), suggesting a potential role for XPO1 inhibitors as a maintenance approach.

Preclinical evidence confirms that selinexor effectively crosses the BBB. In phase II clinical settings, selinexor monotherapy has been shown to induce cell cycle arrest and promote nuclear retention of key TSPs. Early-phase clinical trials have confirmed the feasibility of combining selinexor with temozolomide, with observations of disease stabilization in a small subset of patients, though dose-limiting toxicities remain a primary obstacle (86–90).

In ovarian cancer, early phase I studies indicate that combining selinexor with paclitaxel or PARP inhibitors can elicit responses in platinum-resistant populations. In non-small cell lung cancer, preliminary phase I data suggest that selinexor enhances the efficacy of pembrolizumab by upregulating tumor cell surface expression of MHC class I molecules. In pancreatic ductal adenocarcinoma, the combination of selinexor with gemcitabine or nab-paclitaxel has demonstrated disease control rates in small phase I/II cohorts, though further large-scale validation is required to determine efficacy (91–94).

In summary, although the development of SINE in solid tumors remains at an intermediate level, these agents have already demonstrated measurable biological activity and promising clinical benefit in specific tumor subtypes. Future efforts should focus on identifying robust predictive biomarkers, refining combination regimens and improving toxicity management to advance their translational potential in precision oncology for solid tumors.

The practical deployment of selinexor is considerably restricted by prominent constitutional and hematological toxicities. Rather than being easily managed, these adverse events frequently necessitate immediate treatment interruption, substantial dose reduction or permanent discontinuation, thereby compromising treatment continuity and diminish clinical efficacy.

Thrombocytopenia is the primary dose-limiting hematological toxicity, occurring in up to 43–47% of patients in aggressive schedules (95,96). It requires rigorous weekly monitoring, proactive dose modifications and the implementation of supportive care such as platelet transfusions or thrombopoietin receptor agonists. Neutropenia and anemia are also frequently reported (96,97).

Nausea, vomiting, severe anorexia and fatigue represent profound challenges to patient compliance. Nausea affects >59% of patients, while fatigue affects >61% in some cohorts. To sustain patient adherence, modern dosing protocols have shifted away from initial high-dose, twice-weekly regimens toward once-weekly lower fixed-dose regimens (for example, 60 or 80 mg once weekly) combined with aggressive, multi-agent prophylactic antiemetic regimens (including NK1 receptor antagonists and 5-HT3 antagonists) (72,73,98).

As the therapeutic potential of XPO1 inhibitors continues to evolve, next-generation SINE compounds are under active development to address the pharmacological limitations of selinexor. Eltanexor (KPT-8602) achieves reduced central nervous system toxicity compared with selinexor primarily through structural modifications that lower BBB penetration and shorten target residence time. While eltanexor maintains covalent binding to the Cys528 residue of XPO1, it exhibits modified binding kinetics that result in a shorter duration of target engagement in neuronal cells, thereby sparing the prolonged suppression of nucleocytoplasmic transport key for normal CNS function. Pharmacokinetically, eltanexor was engineered to reduce affinity for the efflux transporter P-glycoprotein (P-gp) at the BBB; this alteration, together with decreased passive permeability, yields notably lower brain-to-plasma ratios than selinexor, limiting drug accumulation in the CNS compartment while preserving effective antitumor concentrations in peripheral tumors and hematopoietic tissues. These molecular and pharmacokinetic features collectively attenuate the neurotoxic effects, such as confusion, ataxia and cognitive disturbance, observed with more CNS-penetrant SINEs and translate clinically into a substantially reduced incidence of severe nausea, vomiting and fatigue. Preclinical and phase I clinical data confirm that eltanexor exerts robust antitumor activity in AML and MM models, offering an improved therapeutic index and greater tolerability for long-term administration. Verdinexor (KPT-335): Initially developed and approved for veterinary oncology in canine lymphoma, verdinexor features high oral bioavailability and extensive peripheral tissue distribution. It is now under evaluation in human preclinical models, demonstrating unique promise in prostate cancer, HIV-associated malignancies and virus-associated malignancies (99–102).

Combination therapy represents the central strategic direction for the future development of selective nuclear export inhibitors. Because XPO1 regulates multiple intersecting signaling networks, its inhibition can enhance the sensitivity of conventional cytotoxic agents, targeted drugs and immunotherapies through distinct synergistic mechanisms.

Chemotherapy sensitization (high clinical maturity) SINE-mediated XPO1 blockade impairs DNA repair pathways, including homologous recombination (HR) and non-homologous end joining (NHEJ), rendering tumor cells more vulnerable to chemotherapy-induced DNA double-strand breaks. Clinically, selinexor combined with cytarabine or anthracyclines has been shown to improve complete remission rates in patients with AML, while its combination with gemcitabine has demonstrated enhanced disease control in pancreatic cancer. Furthermore, a phase I study of selinexor in combination with high-dose cytarabine and mitoxantrone (HiDAC/Mito) achieved an overall response rate of 70% in newly diagnosed or relapsed/refractory AML, confirming that SINE compounds successfully potentiate the efficacy of DNA-damaging agents. In aggressive lymphomas, XPO1-mediated mRNA export of genome maintenance regulators actively drives chemotherapy resistance, making the combination with standard cytotoxic regimens mechanistically sound (74,75).

SINE compounds counteract oncogenic survival signals by blocking the nuclear export of IκBα, leading to nuclear retention of the inhibitor, deactivation of the NF-κB pathway and synergistic induction of apoptosis when paired with bortezomib or carfilzomib. BCL-2 Inhibitors: One of the most promising targeted combinations is the pairing of SINEs with the BCL-2 inhibitor venetoclax in AML. Venetoclax induces mitochondrial apoptosis, but resistance often emerges via MCL-1 upregulation. SINE compounds counteract this resistance by blocking the nuclear export of MCL-1 mRNA via eIF4E-dependent translational regulation, thereby reducing MCL-1 protein levels and restoring sensitivity to venetoclax. The 2026 phase Ib data on Selinexor-venetoclax in relapsed/refractory AML further confirmed the safety and clinical activity of this combination, while the eltanexor-venetoclax trial (NCT06399640) was initiated to extend this strategy to venetoclax-resistant AML populations (76).

PARP Inhibitors: Preclinical models across TNBC, SCLC and glioblastoma demonstrate that combining selinexor with olaparib notably increases anti-tumor activity by simultaneously blocking DNA damage repair and impairing compensatory lysosomal autophagy (55,82,83).

SINEs induce nuclear retention of PD-L1, potentially altering its membrane expression dynamics and sensitizing tumors to PD-1/PD-L1 blockade. In preclinical models, combining selinexor with anti-PD-1 antibodies results in marked tumor growth suppression, though large-scale clinical validation is still pending (77). Dual blockade of nuclear pores: This is a highly innovative approach that simultaneously inhibits nuclear export (through XPO1) and nuclear import (through import proteins α/β). Since inhibiting XPO1 alone may lead to compensatory increases in nuclear import, it is hypothesized that dual targeting can trigger more severe nuclear-cytoplasmic transport disorders, trapping regulatory proteins in the wrong cellular compartments and inducing tumor cell death. This strategy is currently entirely in the preclinical stage (102,103). However, since nuclear-cytoplasmic transport is key for all eukaryotic cells, simultaneously inhibiting XPO1 and import proteins α/β is expected to have an impact on healthy tissues, particularly those with high proliferation rates and active protein transport, such as hematopoietic precursor cells, gastrointestinal epithelial cells and neural progenitor cells. The expected toxic challenges include exacerbated bone marrow suppression (such as thrombocytopenia and neutropenia), cumulative disruption of transcription factor localization in normal cells and potential neurotoxicity resulting from the disruption of nuclear transport in non-dividing neurons. Nevertheless, The present review also notes the differential transport dependence between tumor cells and normal cells, as well as emerging strategies such as tumor-targeted drug delivery and intermittent dosing regimens, which may provide therapeutic windows to mitigate these risks (78,79).

The most robustly documented mechanism of acquired resistance involves specific point mutations within the cargo-binding pocket of XPO1. Prolonged drug exposure drives the selection of malignant clones harboring a cysteine-to-serine mutation at position 528 (C528S). Because SINE compounds rely on forming a covalent bond with the sulfur atom of the Cys528 side chain, this substitution eliminates covalent drug binding while preserving endogenous nuclear export activity, rendering these inhibitors ineffective (87).

Resistant tumor phenotypes frequently display marked transcriptional upregulation of ATP-binding cassette (ABC) transporters, most notably P-gp (encoded by ABCB1/MDR1). Upregulated P-gp accelerates the active efflux of small-molecule SINEs across the plasma membrane, reducing intracellular drug concentrations below the threshold required to achieve effective target occupancy (88,89).

When XPO1-mediated transport is chronically suppressed, tumor cells can undergo adaptive evolution by upregulating alternative, non-canonical nuclear export receptors, such as Exportin-5 (XPO5) and Exportin-7 (XPO7). These alternative transporters take over the nucleocytoplasmic shuttling of vital growth-regulatory mRNAs and tumor suppressors, bypassing XPO1 blockade and restoring the baseline malignant phenotype (90,91).

The absence of validated predictive biomarkers presents a notable obstacle to precise clinical deployment. High XPO1 expression does not necessarily associate with sensitivity to XPO1 inhibition; instead, response is dictated by tumor suppressor status, nuclear export dependency, mutational background and complex RNA regulatory networks (92,93).

Research on tumor prognosis and therapeutic response biomarkers has moved from single-protein expression toward non-coding RNAs, multi-omics features and integrated regulatory networks. This trend is particularly evident in gastrointestinal and colorectal cancer, where numerous clinical evaluations have established that distinct long non-coding RNA and microRNA profiles serve as superior predictive and prognostic indicators compared with traditional immunohistochemistry. Integrating XPO1 biomarker development into this multi-omic landscape offers notable translational opportunities: Synthetic lethality profiling, high-throughput sequencing enables the systematic discovery of genetic alterations, such as TP53 mutations, KRAS aberrations or distinct transcriptomic expression signatures, associated with sensitivity to XPO1 inhibition. For instance, tumors with p53 dysfunction may exhibit heightened dependence on XPO1-mediated survival pathways, rendering them vulnerable to selective inhibition through a synthetic lethal interaction. Dynamic PET Probes: The development of novel positron emission tomography (PET) probes capable of targeting the XPO1 active site represents n achievable near-term clinical realistic goal. Such probes could enable non-invasive, real-time imaging of nuclear export activity and track the subcellular redistribution of TSPs, offering a dynamic approach to patient stratification and early assessment of therapeutic response (94,106–108).

To overcome point mutations (for example, C528S) that render traditional small-molecule inhibitors ineffective, proteolysis-targeting chimera (PROTAC) technology has emerged as a transformative frontier (52,99). Unlike conventional SINEs that merely block protein function, XPO1-directed PROTACs function as bifunctional molecules that recruit specific E3 ubiquitin ligases, such as cereblon (CRBN) or von Hippel-Lindau (VHL), to tag XPO1 for polyubiquitination and subsequent proteasomal degradation. Preclinical proof-of-concept studies demonstrate that PROTAC platforms provide distinct advantages over traditional competitive inhibition (100). Because PROTACs rely on transient, event-driven binding rather than permanent covalent attachment to the Cys528 residue, they retain full degradative capacity against C528S-mutant clones, successfully preventing acquired resistance (101). Furthermore, PROTACs drive the physical destruction of the XPO1 receptor rather than temporary occupancy, inducing more prolonged suppression of nucleocytoplasmic transport and triggering apoptosis at lower concentration thresholds (44). Dual inhibition of XPO1 and cell-cycle regulators has also revealed an acquired vulnerability exploitable by CRBN-based PROTACs in preclinical models (102). Moreover, by optimizing the linker length and selecting tissue-selective E3 ligases, next-generation chimeras can be engineered to minimize degradation within normal hematopoietic precursor cells, potentially mitigating severe clinical toxicities such as thrombocytopenia (103). Importantly, the off-target profiles of PROTACs and covalent inhibitors such as selinexor arise from fundamentally distinct mechanisms: selinexor's dose-limiting toxicities result from systemic, on-target inhibition of XPO1′s physiological export functions across all tissues, whereas PROTACs introduce a unique off-target landscape driven by the recruited E3 ligase, CRBN-based degraders may inadvertently target IKZF1/3 zinc-finger transcription factors, while VHL-recruiting chimeras could perturb the hypoxia-inducible factor pathway (109). However, these risks can be managed by engineering E3 ligase ligands with attenuated neosubstrate activity, fine-tuning linker chemistry to restrict degradation to specific conformational states of XPO1 or exploiting E3 ligases with restricted tissue expression. Therefore, PROTACs effectively decouple therapeutic efficacy from the on-target toxicities inherent to pan-XPO1 covalent inhibition, offering a path toward a more favorable safety profile (44,101).