Efficacy of NEDD8 Pathway Inhibition in Preclinical Models of Poorly Differentiated, Clinically Aggressive Colorectal Cancer

Comprehensive multi-omics analysis elucidates colchicine-induced toxicity mechanisms and unveils the therapeutic potential of MLN4924 and kinase inhibitors

Colchicine is a widely prescribed anti-inflammatory drug for the treatment of gout, familial Mediterranean fever and pericarditis, but its narrow therapeutic window presents a significant risk of severe toxicity. Despite its clinical relevance, the molecular mechanisms underlying colchicine's pharmacological effects and associated toxicity and explored potential therapeutic interventions to mitigate its adverse effects. We showed the colchicine's impact on cellular morphology in human umbilical vein endothelial cells (HUVEC) and HeLa cells including cell rounding and detachment following 24 h of exposure that revealed pronounced cytotoxic effects. We then established a large-scale screening model to identify small molecules capable of reversing colchicine-induced cellular toxicity, and identified MLN4924, an inhibitor of the Cullin-RING E3 ligase (CRL) system, as a promising candidate for mitigating colchicine-induced cellular injury. Through a comprehensive multi-omics approach including transcriptomics, proteomics, phosphoproteomics and ubiquitinomics, we systematically characterized the molecular perturbations caused by colchicine and delineated the protective mechanisms of MLN4924. We found that MLN4924 exerted its protective effects by modulating critical cellular pathways, specifically preventing the dysregulation of cell cycle progression, mitotic disruption and microtubule destabilization triggered by colchicine. Furthermore, proteomic and phosphoproteomic analyses revealed significant alterations in kinase signaling networks, with combined inhibition of CDK1 and PAK1 emerging as an effective strategy to counteract colchicine-induced cellular dysfunction. These results not only provide a detailed molecular characterization of colchicine toxicity but also identify key therapeutic targets, laying the groundwork for the development of targeted interventions to mitigate colchicine-induced adverse effects in clinical practice.

Rab31 promotes metastasis and cisplatin resistance in stomach adenocarcinoma through Twist1-mediated EMT

Stomach adenocarcinoma (STAD) is one of the leading causes of cancer-related death globally. Metastasis and drug resistance are two major causes of failures in current chemotherapy. Here, we found that the expression of Ras-related protein 31 (Rab31) is upregulated in human STAD tissues and high expression of Rab31 is closely associated with poor survival time. Furthermore, we revealed that Rab31 promotes cisplatin resistance and metastasis in human STAD cells. Reduced Rab31 expression induces tumor cell apoptosis and increases cisplatin sensitivity in STAD cells; Rab31 overexpression yielded the opposite result. Rab31 silencing prevented STAD cell migration, whereas the overexpression of Rab31 increased the metastatic potential. Further work showed that Rab31 mediates cisplatin resistance and metastasis via epithelial-mesenchymal transition (EMT) pathway. In addition, we found that both Rab31 overexpression and cisplatin treatment results in increased Twist1 expression. Depletion of Twist1 enhances sensitivity to cisplatin in STAD cells, which cannot be fully reversed by Rab31 overexpression. Rab31 could activate Twist1 by activating Stat3 and inhibiting Mucin 1 (MUC-1). The present study also demonstrates that Rab31 knockdown inhibited tumor growth in mice STAD models. These findings indicate that Rab31 is a novel and promising biomarker and potential therapeutic target for diagnosis, treatment and prognosis prediction in STAD patients. Our data not only identifies a novel Rab31/Stat3/MUC-1/Twist1/EMT pathway in STAD metastasis and drug resistance, but it also provides direction for the exploration of novel strategies to predict and treat STAD in the future.

Case report: Preclinical efficacy of NEDD8 and proteasome inhibitors in patient-derived models of signet ring high-grade mucinous colorectal cancer from a Lynch syndrome patient

High-grade mucinous colorectal cancer (HGM CRC) is particularly aggressive, prone to metastasis and treatment resistance, frequently accompanied by "signet ring" cancer cells. A sizeable fraction of HGM CRCs (20-40%) arises in the context of the Lynch Syndrome, an autosomal hereditary syndrome that predisposes to microsatellite instable (MSI) CRC. Development of patient-derived preclinical models for this challenging subtype of colorectal cancer represents an unmet need in oncology. We describe here successful propagation of preclinical models from a case of early-onset, MSI-positive metastatic colorectal cancer in a male Lynch syndrome patient, refractory to standard care (FOLFOX6, FOLFIRI-Panitumumab) and, surprisingly, also to immunotherapy. Surgical material from a debulking operation was implanted in NOD/SCID mice, successfully yielding one patient-derived xenograft (PDX). PDX explants were subsequently used to generate 2D and 3D cell cultures. Histologically, all models resembled the tumor of origin, displaying a high-grade mucinous phenotype with signet ring cells. For preclinical exploration of alternative treatments, in light of recent findings, we considered inhibition of the proteasome by bortezomib and of the related NEDD8 pathway by pevonedistat. Indeed, sensitivity to bortezomib was observed in mucinous adenocarcinoma of the lung, and we previously found that HGM CRC is preferentially sensitive to pevonedistat in models with low or absent expression of cadherin 17 (CDH17), a differentiation marker. We therefore performed IHC on the tumor and models, and observed no CDH17 expression, suggesting sensitivity to pevonedistat. Both bortezomib and pevonedistat showed strong activity on 2D cells at 72 hours and on 3D organoids at 7 days, thus providing valid options for in vivo testing. Accordingly, three PDX cohorts were treated for four weeks, respectively with vehicle, bortezomib and pevonedistat. Both drugs significantly reduced tumor growth, as compared to the vehicle group. Interestingly, while bortezomib was more effective in vitro, pevonedistat was more effective in vivo. Drug efficacy was further substantiated by a reduction of cellularity and of Ki67-positive cells in the treated tumors. These results highlight proteasome and NEDD8 inhibition as potentially effective therapeutic approaches against Lynch syndrome-associated HGM CRC, also when the disease is refractory to all available treatment options.

Inhibition of Neddylation Suppresses Osteoclast Differentiation and Function In Vitro and Alleviates Osteoporosis In Vivo

Neddylation, or the covalent addition of NEDD8 to specific lysine residue of proteins, is a reversible posttranslational modification, which regulates numerous biological functions; however, its involvement and therapeutic significance in osteoporosis remains unknown. Our results revealed that during the soluble receptor activator of nuclear factor-κB ligand (sRANKL)-stimulated osteoclast differentiation, the neddylation and expression of UBA3, the NEDD8-activating enzyme (NAE) catalytic subunit, were dose- and time-dependently upregulated in RAW 264.7 macrophages. UBA3 knockdown for diminishing NAE activity or administering low doses of the NAE inhibitor MLN4924 significantly suppressed sRANKL-stimulated osteoclast differentiation and bone-resorbing activity in the macrophages by inhibiting sRANKL-stimulated neddylation and tumor necrosis factor receptor-associated factor 6 (TRAF6)-activated transforming growth factor-β-activated kinase 1 (TAK1) downstream signaling for diminishing nuclear factor-activated T cells c1 (NFATc1) expression. sRANKL enhanced the interaction of TRAF6 with the neddylated proteins and the polyubiquitination of TRAF6's lysine 63, which activated TAK1 downstream signaling; however, this process was inhibited by MLN4924. MLN4924 significantly reduced osteoporosis in an ovariectomy- and sRANKL-induced osteoporosis mouse model in vivo. Our novel finding was that NAE-mediated neddylation participates in RANKL-activated TRAF6-TAK1-NFATc1 signaling during osteoclast differentiation and osteoporosis, suggesting that neddylation may be a new target for treating osteoporosis.

SOMCL-19-133, a novel, selective, and orally available inhibitor of NEDD8-activating enzyme (NAE) for cancer therapy

Inhibition of the NEDD8-activating enzyme (NAE), the key E1 enzyme in the neddylation cascade, has been considered an attractive anticancer strategy with the discovery of the first-in-class NAE inhibitor, MLN4924. In this study, we identified SOMCL-19-133 as a highly potent, selective, and orally available NAE inhibitor, which is an analog to AMP. It effectively inhibited NAE with an IC₅₀ value of 0.36 nM and exhibited more than 2855-fold selectivity over the closely related Ubiquitin-activating enzyme (UAE). It is worth noting that treatment with SOMCL-19-133 prominently inhibited Cullin neddylation and delayed the turnover of a panel of Cullin-RING ligases (CRLs) substrates (e.g., Cdt1, p21, p27, and Wee1) at lower effective concentrations than that of MLN4924, subsequently caused DNA damage and Chk1/Chk2 activation, and thus triggered cell cycle arrest and apoptosis. Moreover, SOMCL-19-133 exhibited potent antiproliferative activity against a broad range of human tumor cell lines (mean IC₅₀ 201.11 nM), which was about 5.31-fold more potent than that of MLN4924. In vivo, oral delivery treatments with SOMCL-19-133, as well as the subcutaneous injection, led to significant tumor regression in mouse xenograft models. All of the treatments were well tolerated on a continuous daily dosing schedule. Compared with MLN4924, SOMCL-19-133 had a 5-fold higher peak plasma concentration, lower plasma clearance, and a 4-fold larger area under the curve (AUC_last). In conclusion, SOMCL-19-133 is a promising preclinical candidate for treating cancers owing to its profound in vitro and in vivo efficacy and favorable pharmacokinetic properties.

The development of pevonedistat in myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML): hope or hype?

Myelodysplastic syndrome (MDS) is a clonal hematopoietic stem cell disorder clinically defined by cytopenias, bone marrow failure, and an increased risk of progressing to acute myeloid leukemia (AML). Traditionally, first-line treatment for patients with higher-risk MDS has been hypomethylating agents (HMAs). However, these agents have modest clinical activity as single agents. A one-size-fits-all treatment paradigm is insufficient for such a heterogeneous disease in the modern era of precision medicine. Several new agents have been developed for MDS with the hopes of improving clinical outcomes and survival. Pevonedistat is a first-in-class, novel inhibitor of neuronal precursor cell-expressed developmentally down-regulated protein-8 (NEDD8) activating enzyme (NAE) blocking the neddylation pathway leading to downstream effects on the ubiquitin-proteosome pathway. Pevonedistat ultimately leads to apoptosis and inhibition of the cell cycle in cancer cells. Studies have demonstrated the safety profile of pevonedistat, leading to the development of multiple trials investigating combination strategies with pevonedistat in MDS and AML. In this review, we summarize the preclinical and clinical rationale for pevonedistat in MDS and AML, review the clinical data of this agent alone and in combination with HMAs to date, and highlight potential future directions for this agent in myeloid malignancies.

SKP2 drives the sensitivity to neddylation inhibitors and cisplatin in malignant pleural mesothelioma

Background

The combination of pemetrexed and cisplatin remains the reference first-line systemic therapy for malignant pleural mesothelioma (MPM). Its activity is moderate because of tumor aggressiveness, immune-suppressive environment and resistance to chemotherapy-induced immunogenic cell death (ICD). Preliminary and limited findings suggest that MPM cells have deregulated ubiquitination and proteasome activities, although proteasome inhibitors achieved disappointing clinical results.

Methods

Here, we investigated the role of the E3-ubiquitin ligase SKP/Cullin/F-box (SCF) complex in cell cycle progression, endoplasmic reticulum (ER)/proteostatic stress and ICD in MPM, and the therapeutic potential of the neddylation/SCF complex inhibitor MLN4924/Pevonedistat.

Results

In patient-derived MPM cultures and syngenic murine models, MLN4924 and cisplatin showed anti-tumor effects, regardless of MPM histotype and BAP1 mutational status, increasing DNA damage, inducing S- and G2/M-cell cycle arrest, and apoptosis. Mechanistically, by interfering with the neddylation of cullin-1 and ubiquitin-conjugating enzyme UBE2M, MLN4924 blocks the SCF complex activity and triggers an ER stress-dependent ICD, which activated anti-MPM CD8⁺T-lymphocytes. The SKP2 component of SCF complex was identified as the main driver of sensitivity to MLN4924 and resistance to cisplatin. These findings were confirmed in a retrospective MPM patient series, where SKP2 high levels were associated with a worse response to platinum-based therapy and inferior survival.

Conclusions

We suggest that the combination of neddylation inhibitors and cisplatin could be worth of further investigation in the clinical setting for MPM unresponsive to cisplatin. We also propose SKP2 as a new stratification marker to determine the sensitivity to cisplatin and drugs interfering with ubiquitination/proteasome systems in MPM.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13046-022-02284-7.

Synthetic Lethality Screening Highlights Colorectal Cancer Vulnerability to Concomitant Blockade of NEDD8 and EGFR Pathways

Simple Summary

Identification of effective therapies for clinically aggressive, treatment-resistant colorectal cancer (CRC) remains an unmet clinical need. Targeted therapies against the epidermal growth factor receptor (EGFR) signaling axis lead to clinical benefits only in a small fraction of patients due to primary and acquired resistance. We previously showed that the NEDD8 pathway inhibitor pevonedistat induced tumor stabilization in preclinical models of aggressive CRC. Here, through synthetic lethality screenings, we found that pevonedistat could be successfully combined with EGFR pathway-targeted treatments in BRAF-mutant and RAS-RAF wild-type CRCs originally resistant to BRAF and EGFR blockade. We found that combined blockade of NEDD8 and EGFR pathways reverted compensatory feedback loops that reduced the efficacy of single treatments. Our results provide preclinical validation of a promising therapeutic strategy for clinically aggressive CRC resistant to EGFR and BRAF-targeted treatments.

Abstract

Colorectal cancer (CRC) is a heterogeneous disease showing significant variability in clinical aggressiveness. Primary and acquired resistance limits the efficacy of available treatments, and identification of effective drug combinations is needed to further improve patients’ outcomes. We previously found that the NEDD8-activating enzyme inhibitor pevonedistat induced tumor stabilization in preclinical models of poorly differentiated, clinically aggressive CRC resistant to available therapies. To identify drugs that can be effectively combined with pevonedistat, we performed a “drop-out” loss-of-function synthetic lethality screening with an shRNA library covering 200 drug-target genes in four different CRC cell lines. Multiple screening hits were found to be involved in the EGFR signaling pathway, suggesting that, rather than inhibition of a specific gene, interference with the EGFR pathway at any level could be effectively leveraged for combination therapies based on pevonedistat. Exploiting both BRAF-mutant and RAS/RAF wild-type CRC models, we validated the therapeutic relevance of our findings by showing that combined blockade of NEDD8 and EGFR pathways led to increased growth arrest and apoptosis both in vitro and in vivo. Pathway modulation analysis showed that compensatory feedback loops induced by single treatments were blunted by the combinations. These results unveil possible therapeutic opportunities in specific CRC clinical settings.

Preclinical models as patients' avatars for precision medicine in colorectal cancer: past and future challenges

Colorectal cancer (CRC) is a complex and heterogeneous disease, characterized by dismal prognosis and low survival rate in the advanced (metastatic) stage. During the last decade, the establishment of novel preclinical models, leading to the generation of translational discovery and validation platforms, has opened up a new scenario for the clinical practice of CRC patients. To bridge the results developed at the bench with the medical decision process, the ideal model should be easily scalable, reliable to predict treatment responses, and flexibly adapted for various applications in the research. As such, the improved benefit of novel therapies being tested initially on valuable and reproducible preclinical models would lie in personalized treatment recommendations based on the biology and genomics of the patient's tumor with the overall aim to avoid overtreatment and unnecessary toxicity. In this review, we summarize different in vitro and in vivo models, which proved efficacy in detection of novel CRC culprits and shed light into the biology and therapy of this complex disease. Even though cell lines and patient-derived xenografts remain the mainstay of colorectal cancer research, the field has been confidently shifting to the use of organoids as the most relevant preclinical model. Prioritization of organoids is supported by increasing body of evidence that these represent excellent tools worth further therapeutic explorations. In addition, novel preclinical models such as zebrafish avatars are emerging as useful tools for pharmacological interrogation. Finally, all available models represent complementary tools that can be utilized for precision medicine applications.

Nicotine-derived NNK induces the stemness enrichment of CRC cells through regulating the balance of DUSP4-ERK1/2 feedback loop

Cigarette smoking has been considered as an independent risk factor for colorectal cancer (CRC) initiation and progression. In this study, we found that cigarette smoking was significantly associated with poor CRC differentiation (P = 0.040). Since studies have indicated that poorly differentiated tumors are more aggressive and metastasize earlier, leading to poorer prognosis; and cancer stem cells (CSCs) are largely responsible for tumor differentiation state, here we observed that the exposure of nicotine-derived 4-(methylnitrosamino)- 1-(3-pyridyl)- 1-butanone (NNK) promoted cell sphere formation and the expression of the stem cell markers, CD44, OCT4, C-MYC and NANOG in HCT8 and DLD-1 cells. Further colony formation assay, CCK-8 assay and tumor-bearing experiment showed that NNK exposure significantly increased the proliferative and growth ability of CRC cells. In mechanism, we found that NNK-activated ERK1/2 played an important role in enrichment of CRC stem cells and the up-regulation of DUSP4, a major negative regulator of ERK1/2. Moreover, DUSP4 up-regulation was essential for maintaining NNK-activated ERK1/2 in an appropriate level, which was an required event for NNK-induced stemness enrichment of CRC cells. Taken together, our findings provided a possible mechanistic insight into cigarette smoking-induced CRC progression.

E1 Enzymes as Therapeutic Targets in Cancer

Post-translational modifications of cellular substrates with ubiquitin and ubiquitin-like proteins (UBLs), including ubiquitin, SUMOs, and neural precursor cell-expressed developmentally downregulated protein 8, play a central role in regulating many aspects of cell biology. The UBL conjugation cascade is initiated by a family of ATP-dependent enzymes termed E1 activating enzymes and executed by the downstream E2-conjugating enzymes and E3 ligases. Despite their druggability and their key position at the apex of the cascade, pharmacologic modulation of E1s with potent and selective drugs has remained elusive until 2009. Among the eight E1 enzymes identified so far, those initiating ubiquitylation (UBA1), SUMOylation (SAE), and neddylation (NAE) are the most characterized and are implicated in various aspects of cancer biology. To date, over 40 inhibitors have been reported to target UBA1, SAE, and NAE, including the NAE inhibitor pevonedistat, evaluated in more than 30 clinical trials. In this Review, we discuss E1 enzymes, the rationale for their therapeutic targeting in cancer, and their different inhibitors, with emphasis on the pharmacologic properties of adenosine sulfamates and their unique mechanism of action, termed substrate-assisted inhibition. Moreover, we highlight other less-characterized E1s-UBA6, UBA7, UBA4, UBA5, and autophagy-related protein 7-and the opportunities for targeting these enzymes in cancer. SIGNIFICANCE STATEMENT: The clinical successes of proteasome inhibitors in cancer therapy and the emerging resistance to these agents have prompted the exploration of other signaling nodes in the ubiquitin-proteasome system including E1 enzymes. Therefore, it is crucial to understand the biology of different E1 enzymes, their roles in cancer, and how to translate this knowledge into novel therapeutic strategies with potential implications in cancer treatment.

Pevonedistat (MLN4924): mechanism of cell death induction and therapeutic potential in colorectal cancer

Pevonedistat (MLN4924), a selective inhibitor of the NEDD8-activating enzyme E1 regulatory subunit (NAE1), has demonstrated significant therapeutic potential in several malignancies. Although multiple mechanisms-of-action have been identified, how MLN4924 induces cell death and its potential as a combinatorial agent with standard-of-care (SoC) chemotherapy in colorectal cancer (CRC) remains largely undefined. In an effort to understand MLN4924-induced cell death in CRC, we identified p53 as an important mediator of the apoptotic response to MLN4924. We also identified roles for the extrinsic (TRAIL-R2/caspase-8) and intrinsic (BAX/BAK) apoptotic pathways in mediating the apoptotic effects of MLN4924 in CRC cells, as well as a role for BID, which modulates a cross-talk between these pathways. Depletion of the anti-apoptotic protein FLIP, which we identify as a novel mediator of resistance to MLN4924, enhanced apoptosis in a p53-, TRAIL-R2/DR5-, and caspase-8-dependent manner. Notably, TRAIL-R2 was involved in potentiating the apoptotic response to MLN4924 in the absence of FLIP, in a ligand-independent manner. Moreoever, when paired with SoC chemotherapies, MLN4924 demonstrated synergy with the irinotecan metabolite SN38. The cell death induced by MLN4924/SN38 combination was dependent on activation of mitochondria through BAX/BAK, but in a p53-independent manner, an important observation given the high frequency of TP53 mutation(s) in advanced CRC. These results uncover mechanisms of cell death induced by MLN4924 and suggest that this second-generation proteostasis-disrupting agent may have its most widespread activity in CRC, in combination with irinotecan-containing treatment regimens.

Proteome Instability Is a Therapeutic Vulnerability in Mismatch Repair-Deficient Cancer

Deficient DNA mismatch repair (dMMR) induces a hypermutator phenotype that can lead to tumorigenesis; however, the functional impact of the high mutation burden resulting from this phenotype remains poorly explored. Here, we demonstrate that dMMR-induced destabilizing mutations lead to proteome instability in dMMR tumors, resulting in an abundance of misfolded protein aggregates. To compensate, dMMR cells utilize a Nedd8-mediated degradation pathway to facilitate clearance of misfolded proteins. Blockade of this Nedd8 clearance pathway with MLN4924 causes accumulation of misfolded protein aggregates, ultimately inducing immunogenic cell death in dMMR cancer cells. To leverage this immunogenic cell death, we combined MLN4924 treatment with PD1 inhibition and found the combination was synergistic, significantly improving efficacy over either treatment alone.

Current and Future Horizons of Patient-Derived Xenograft Models in Colorectal Cancer Translational Research

Our poor understanding of the intricate biology of cancer and the limited availability of preclinical models that faithfully recapitulate the complexity of tumors are primary contributors to the high failure rate of novel therapeutics in oncology clinical studies. To address this need, patient-derived xenograft (PDX) platforms have been widely deployed and have reached a point of development where we can critically review their utility to model and interrogate relevant clinical scenarios, including tumor heterogeneity and clonal evolution, contributions of the tumor microenvironment, identification of novel drugs and biomarkers, and mechanisms of drug resistance. Colorectal cancer (CRC) constitutes a unique case to illustrate clinical perspectives revealed by PDX studies, as they overcome limitations intrinsic to conventional ex vivo models. Furthermore, the success of molecularly annotated "Avatar" models for co-clinical trials in other diseases suggests that this approach may provide an additional opportunity to improve clinical decisions, including opportunities for precision targeted therapeutics, for patients with CRC in real time. Although critical weaknesses have been identified with regard to the ability of PDX models to predict clinical outcomes, for now, they are certainly the model of choice for preclinical studies in CRC. Ongoing multi-institutional efforts to develop and share large-scale, well-annotated PDX resources aim to maximize their translational potential. This review comprehensively surveys the current status of PDX models in translational CRC research and discusses the opportunities and considerations for future PDX development.

Disrupting G6PD-mediated Redox homeostasis enhances chemosensitivity in colorectal cancer

Glucose-6-phosphate dehydrogenase (G6PD) is a key enzyme that generates NADPH to maintain reduced glutathione (GSH), which scavenges reactive oxygen species (ROS) to protect cancer cell from oxidative damage. In this study, we mainly investigate the potential roles of G6PD in colorectal cancer (CRC) development and chemoresistance. We discover that G6PD is overexpressed in CRC cells and patient specimens. High expression of G6PD predicts poor prognosis and correlated with poor outcome of oxaliplatin-based first-line chemotherapy in patients with CRC. Suppressing G6PD decreases NADPH production, lowers GSH levels, impairs the ability to scavenge ROS levels, and enhances oxaliplatin-induced apoptosis in CRC via ROS-mediated damage in vitro. In vivo experiments further shows that silencing G6PD with lentivirus or non-viral gene delivery vector enhances oxaliplatin anti-tumor effects in cell based xenografts and PDX models. In summary, our finding indicated that disrupting G6PD-mediated NADPH homeostasis enhances oxaliplatin-induced apoptosis in CRC through redox modulation. Thus, this study indicates that G6PD is a potential prognostic biomarker and a promising target for CRC therapy.

Loss of AXIN1 drives acquired resistance to WNT pathway blockade in colorectal cancer cells carrying RSPO3 fusions

In colorectal cancer (CRC), WNT pathway activation by genetic rearrangements of RSPO3 is emerging as a promising target. However, its low prevalence severely limits availability of preclinical models for in-depth characterization. Using a pipeline designed to suppress stroma-derived signal, we find that RSPO3 "outlier" expression in CRC samples highlights translocation and fusion transcript expression. Outlier search in 151 CRC cell lines identified VACO6 and SNU1411 cells as carriers of, respectively, a canonical PTPRK(e1)-RSPO3(e2) fusion and a novel PTPRK(e13)-RSPO3(e2) fusion. Both lines displayed marked in vitro and in vivo sensitivity to WNT blockade by the porcupine inhibitor LGK974, associated with transcriptional and morphological evidence of WNT pathway suppression. Long-term treatment of VACO6 cells with LGK974 led to the emergence of a resistant population carrying two frameshift deletions of the WNT pathway inhibitor AXIN1, with consequent protein loss. Suppression of AXIN1 in parental VACO6 cells by RNA interference conferred marked resistance to LGK974. These results provide the first mechanism of secondary resistance to WNT pathway inhibition.