Targeted protein degradation of Wnt/β-catenin signaling pathway: an effective strategy for cancer therapy

Abstract

The Wnt/β-Catenin signaling pathway is highly conserved and initiated by a multiprotein signalosome complex. It is essential for embryonic development, organ formation, and tissue homeostasis. Abnormal Wnt signaling is found in many malignant tumors, and Wnt mutations are widely known to drive tumor initiation. Since the discovery of its oncogenic roles, numerous therapeutic agents have been developed, but none have been approved. This is attributed to the inherent structural characteristics of its core components and their role in normal tissue homeostasis. The majority of clinical drugs are occupancy-driven small molecule inhibitors, and most encounter challenges such as off-target effects and adverse toxicities. In recent years, targeted protein degradation (TPD) technology has emerged as a promising approach, offering a novel strategy for drug development targeting the Wnt/β-Catenin pathway. Utilizing this technology to degrade key proteins in the Wnt/β-Catenin pathway holds significant potential for inhibiting its oncogenic effects and restoring the cancer-immunity cycle (CI cycle). This review presents a systematic analysis of the “undruggability” of the Wnt/β-Catenin pathway and proposes potential strategies ranging from target selection to TPD application, with the aim of offering novel insights into overcoming this challenge.

Introduction

The canonical Wingless/Int-1 (Wnt) signaling pathway, commonly referred to as the Wnt/β-Catenin signaling pathway, crucially depends on the activation of a multiprotein signalosome complex formed by Wnt ligands (Wnts), Frizzled receptors (Fzds), and low-density lipoprotein receptor-related protein 5 and 6 (LRP5/6) co-receptors. Upon activation, this complex triggers the nuclear translocation of β-catenin, which subsequently binds to T cell factor/lymphoid enhancer-binding factor (TCF/LEF) transcription factors and activates downstream target genes [1, 2]. Moreover, the Wnt/β-Catenin pathway exhibits extensive crosstalk with other conserved pathways, forming a complicated network to orchestrate multiple cellular functions [3, 4].

The canonical Wnt pathway is highly conserved and mediates a wide range of biological functions across both physiological and pathological contexts [1–6]. The aberrantly activated Wnt/β-Catenin signaling pathway has become a master regulatory axis in oncogenesis [1–4]. On the one hand, hyperactivation and mutations of the Wnt pathway in cancer cells induce the expression of downstream malignant target genes such as c-MYC, vascular endothelial growth factor (VEGF), CD44, and matrix metalloproteinases (MMPs), thereby leading to high tumor heterogeneity [4, 7]. On the other hand, Wnt signaling regulates the types and proportions of immune cells in the tumor microenvironment (TME) [7, 8], and maintains the phenotype of cancer-associated fibroblasts (CAFs) in the extracellular matrix (ECM) [9]. As a result, this regulation contributes to the formation of immune-excluded and immune-desert TMEs, which give rise to “cold tumors” [10, 11]. Finally, the reprogrammed TME further disrupts the cancer-immunity cycle (CI cycle) and gives rise to issues such as drug resistance [12].

Despite playing a central role in oncopathology, the Wnt/β-Catenin pathway has long been categorized as an “undruggable” pathway due to various pharmacological challenges. The majority of Wnt-targeting drugs under clinical investigation are designed to modulate protein activity, among which inhibitors account for a significant proportion [13]. These agents primarily focus on disrupting the formation of the multiprotein signalosome complex, enhancing β-catenin degradation by the destruction complex (DC), and directly inhibiting β-catenin or its binding to nuclear co-factors. Nonetheless, considering the essential role of Wnt/β-Catenin pathway in normal cellular functions [3], the development of traditional targeting agents has been restricted by concerns about potential toxicity and off-target effects. Additionally, the absence of distinct binding pockets in β-catenin and other targets poses significant challenges to the development of targeted therapies [14]. Consequently, no drugs specifically targeting the Wnt/β-Catenin pathway have been approved.

In recent years, novel drug designs have been advancing from bench to bedside. TPD technologies, which specifically include both ubiquitin-proteasome system (UPS)-dependent approaches and lysosome-mediated degradation pathways, are at the forefront of this transformation [15]. Compared to traditional occupancy-driven small molecule inhibitors, TPD drugs directly induce the degradation of target proteins, thereby eliminating their functions. This approach not only effectively mitigates adverse side effects but also circumvents acquired drug resistance that may arise from the protein mutations.

This review systematically analyzes the recent advances as well as persistent challenges of Wnt-targeting traditional inhibitors. The reasons for “undruggability” are summarized, potential strategies to surmount these challenges are proposed, and the breakthroughs achieved by TPD therapeutics in the canonical Wnt pathway are detailed. Ultimately, this review aims not only to offer a clear understanding of the Wnt/β-Catenin signaling pathway but also to inspire new strategies for unlocking the “black box” of “undruggability” in this pathway.

β-Catenin dependent cascade transduction

The Wnt/β-Catenin pathway primarily relies on the cascade transduction of β-catenin (Fig. 1). Wnt pathway activation requires Wnts, which are acylated by the O-acetyltransferase Porcupine (PORCN) [16]. In the absence of Wnt binding, β-catenin undergoes continual degradation by the DC composed of adenomatous polyposis coli (APC), axis inhibition protein (AXIN), glycogen synthase kinase 3β (GSK3β), and casein kinase 1α (CK1α). This complex initially phosphorylates β-catenin, which is then ubiquitinated by the E3 ubiquitin ligase SKP1-CUL1-F-box protein-β-transducin repeat-containing protein (β-TrCP) and ultimately degraded via the proteasomal pathway [16–18].

Fig. 1.

Wnt/β-Catenin pathway relies on the cascade transduction of β-catenin. Activation of the canonical Wnt pathway requires the formation of a multiprotein signalosome complex, which includes Wnts, Fzds, and LRP5/6. This complex initiates a signaling cascade that promotes the nuclear translocation of β-catenin. Once in the nucleus, β-catenin binds to TCF/LEF transcription factors and associated coactivators, thereby activating the expression of downstream target genes. In the absence of Wnts binding, the pathway remains Wnt off, and β-catenin undergoes continuous degradation by the destruction complex, which consists of APC, AXIN1, GSK3β, and CK1α. This complex phosphorylates β-catenin, marking it for ubiquitination by the E3 ubiquitin ligase β-TrCP and subsequent degradation via the proteasomal pathway

Upon the binding of Wnts to Fzds and LRP5/6, the Dishevelled (DVL) protein is recruited to the membrane, facilitating multimerization of the receptor complex and phosphorylation of GSK3β, thereby inhibiting its activity. Consequently, unphosphorylated β-catenin accumulates in the cytoplasm and translocates into the nucleus, where it binds to TCF/LEF and other coactivators to promote specific changes in the transcriptional machinery.

Conversely, the canonical Wnt pathway is stringently regulated by extracellular repressor proteins. Ring finger protein 43 (RNF43) and zinc and ring finger protein 3 (ZNRF3) are transmembrane E3 ubiquitin ligases (Fig. 1). They regulate Wnt signaling by ubiquitinating Fzds and facilitating their rapid endocytosis and subsequent lysosomal degradation [19, 20]. In contrast, R-spondins (RSPO1-4) enhance Wnt signaling. RSPOs bind to leucine-rich repeat-containing G protein-coupled receptor (LGR4-6) and form complexes with RNF43/ZNRF3, leading to the removal of these complexes [19–23]. Additionally, RSPOs can activate the Wnt/β-Catenin pathway independently of LGRs by interacting with heparan sulfate proteoglycans (HSPGs) [24].

The intricate functions of the Wnt/β-Catenin signaling pathway: challenges in therapeutic targeting

The Wnt/β-Catenin signaling pathway is highly conserved and serves as a central regulator of embryogenesis, organogenesis, and stem cell maintenance [3, 25–29]. Specifically, Wnt signaling regulates osteoblast differentiation and regeneration to maintain bone homeostasis [30–32]. In the liver, Wnt signaling promotes hepatocyte proliferation, aiding tissue repair and functional recovery [33–35]. In the skin, it controls hair follicle cycling and epidermal stem cell activation [36]. Beyond its roles in normal physiology, the Wnt pathway is regulated by various natural inhibitors, classified as secreted or transmembrane proteins (Fig. 2A), and engages in extensive crosstalk with other conserved pathways, forming complex regulatory networks (Fig. 2B). These diverse functions highlight the essential role of Wnt signaling in development and tissue homeostasis, and also underscore the signaling complexity that must be considered when targeting this pathway for developing therapeutics.

Fig. 2.

Canonical Wnt signaling pathway is tightly regulated by a variety of natural inhibitors and interacts with multiple signaling pathways. A Canonical Wnt pathway is modulated by diverse natural inhibitors categorized as secreted or transmembrane types. Secreted inhibitors include Dkk, SOST, Wise, IGFBP-4, Bighead, sFRPs, WIF-1, and CER. Transmembrane inhibitors include Shisa, Tiki, APCDD1, and Waif1/5T4. B Overview diagram of the crosstalk between the canonical Wnt pathway and other pathways, including Notch, Hedgehog, TGF-β, FGF, NF-κB, Hippo, and STAT3

However, abnormally activated and mutated Wnt signaling not only promotes cancer progression by altering tumor cell phenotypes but also further hinders the CI cycle by remodeling the TME to create a “greenhouse” for tumor cells (Fig. 3). We aim to gain deeper insights into the oncogenic mechanisms of aberrant Wnt signaling by analyzing its roles in both tumor cells and the TME.

Fig. 3.

Wnt/β-Catenin signaling promotes oncogenesis and disrupts antitumor immune response. β-catenin mediates immunosuppression by promoting expansion of Tregs, impairing antigen presentation, and inducing ATF3-mediated dysfunction in CD103⁺ cDC1s, ultimately inhibiting CD8⁺ T cell proliferation and migration. Canonical Wnt signaling enhances tumor cell dissemination through the inducing of EMT, CSCs activation, transformation of CAFs, and stimulation of angiogenesis. Concurrently, Wnt-driven upregulation of PD-L1 facilitates tumor immune escape via T cell exhaustion

Wnt/β-Catenin signaling pathway in cancer cells

The activation of proto-oncogenes and inactivation of tumor suppressor genes often serve as initiating events in oncogenesis. Specifically, canonical Wnt signaling promotes oncogenic transformation of normal cells through aberrant activation and mutations. CTNNB1, a proto-oncogene encoding β-catenin, is frequently mutated across multiple cancers [37–44]. Missense mutations affecting serine/threonine residues in exon 3 of CTNNB1 or within ubiquitination-destruction motifs impair the phosphorylation of β-catenin. This impairment leads to its stabilization and nuclear translocation, resulting in uncontrolled transcriptional activation [40, 41]. Approximately half of hepatocellular carcinoma (HCC) patients exhibit activated Wnt signaling along with mutations in CTNNB1, AXIN, and APC [39]. HCC cases harboring gain-of-function (GOF) mutations in CTNNB1 typically present as “cold tumors”, which are associated with poor responses to immunotherapy [43, 44]. Furthermore, CTNNB1 has been shown to induce somatic mutations in gastric epithelial cells [45].

Inactivating mutations in components of the DC responsible for β-catenin degradation represent another major oncogenic mechanism [46]. Loss-of-function (LOF) mutations in APC are a key driver in most colorectal cancers (CRCs). Genetic alterations in APC result in truncated protein products that activate the Wnt signaling pathway and disrupt multiple cellular processes. Deletion of the C-terminal sequence of APC is critical for CRC development due to the loss of APC’s tumor-suppressive function. Moreover, APC deficiency activates NOTUM, an extracellular palmitoylated protein carboxylesterase and a feedback inhibitor of canonical Wnt signaling [47, 48]. AXIN mutations have been identified in multiple cancers, including HCC, Uterine Corpus Endometrial Carcinoma (UCEC), and CRC [46]. Notably, AXIN1 mutations are frequently associated with loss of heterozygosity (LOH) [49]. Recent studies indicate that missense driver mutations in AXIN1 structurally destabilize the RGS domain and mediate the formation of small aggregates that rewire the AXIN1 interactome, thereby promoting basal Wnt pathway activation [46]. However, the functional consequences of most AXIN1 missense mutations remain unclear.

RNF43 and ZNRF3 act as negative regulators of Wnt signaling and are frequently mutated in numerous cancers [46, 50]. RNF43 mutations predominantly include truncating and missense variants, whereas ZNRF3 mutations predominantly involve missense mutations and deletions. Recent findings show that 27 out of 82 ZNRF3 variants located within the RING and RSPO domains cause LOF effects. Additionally, hyperactivating missense mutations in both RNF43 and ZNRF3 may display oncoprotein-like activity [51].

Furthermore, downstream Wnt proto-oncogenes such as c-MYC and cyclin D1 also play pivotal roles in oncogenesis [44]. Approximately 28% of tumors demonstrate MYC overexpression, with upregulated c-MYC serving as a key driver of malignant cellular proliferation. Wnt signaling enhances MYC mRNA transport by modulating CTCF binding sites (CTCFBS) within oncogenic super-enhancers (OSEs) [52–55]. In Wnt-activated cancer cells, cyclin D1 expression is elevated, thereby promoting tumor cell proliferation [56].

Cancer stem cells (CSCs) possess self-renewal and sustained proliferative capacities, contributing significantly to tumor initiation, metastasis, and heterogeneity. The Wnt/β-Catenin pathway regulates cancer stemness across diverse malignancies [57]. For example, in glioblastoma multiforme (GBM), Wnt/β-Catenin-mediated upregulation of neuroligin 3 (NLGN3) increases NLGN3 secretion into the TME, reprogramming neighboring cells into CSCs [58]. In triple-negative breast cancer (TNBC), hypoxia-induced Wnt pathway activation expands the TNBCSCs subpopulation and promotes shedding of cell surface MHC class I-related chain A (MICA), thereby reducing NK cell cytotoxicity [59]. In cervical cancer, LGR6 activates Transcription Factor 7-Like 2 (TCF7L2) via Wnt/β-Catenin signaling. TCF7L2 then forms a complex with nuclear β-catenin and binds to the LGR6 promoter, enhancing LGR6 transcription and forming a positive feedback loop that amplifies Wnt signaling [60].

Cancer metastasis is coordinated by Wnt signaling through epithelial-mesenchymal transition (EMT) and CSC regulation [56]. EMT results in the loss of epithelial polarity, reduced intercellular adhesion, and increased invasiveness and migratory capacity [61–63]. Wnt activation downregulates E-cadherin while upregulating N-cadherin, vimentin, Snail, Slug, and MMP9. Accumulating evidence indicates that EMT enriches for cells with CSC properties, which are considered the cellular origin of cancer progression. For instance, in CRC, Transmembrane 4 L6 Family Member 1 (TM4SF1) promotes metastasis and sustains EMT phenotypes and tumor stemness via the Wnt/β-Catenin/c-MYC/SOX2 axis [64]. Similarly, Runt-related Transcription Factor 1 (RUNX1) activates Wnt/β-Catenin signaling in CRC cells by interacting with β-catenin and targeting the KIT promoter/enhancer regions, thereby driving metastasis and EMT [65]. YTH Domain Family Member 2 (YTHDF2) enhances GBM migration, invasion, and EMT by binding and degrading mRNAs of APC and GSK3β [66]. Collectively, tumor metastasis is governed by multiple factors, with canonical Wnt pathway activation accelerating this process through CSC expansion and EMT induction.

Wnt/β-Catenin signaling pathway in TME

The TME is a highly structured ecosystem comprising immune cells, CAFs, the lymphovascular system, and extracellular components such as ECM, basement membrane, growth factors, and metabolites [67]. The interaction between tumors and the TME is critical for sustaining tumor growth, promoting metastasis, and enhancing malignancy. Wnt promotes the formation of a tumor “greenhouse” by interacting with various components of the TME.

Wnt/β-Catenin signaling pathway and immunity

Over the past decade, immune checkpoint inhibitors (ICIs) have shifted cancer therapy from conventional chemotherapy to immuno-oncology (IO) [68, 69]. However, immune cell infiltration in the TME is now recognized as a key determinant of the immunotherapy response. Clinically, “cold tumors” are frequently associated with the activation of the Wnt/β-Catenin pathway [11, 70, 71]. Consistent with this observation, analysis of The Cancer Genome Atlas (TCGA) revealed that the mutation enrichment of Wnt molecules in non-T cell-inflamed tumor tissues was threefold higher than that in T cell-inflamed tumors. Furthermore, activation of Wnt/β-Catenin signaling is negatively correlated with the expression of T inflammation-related genes in 90% of cancers [72–74]. Building on these findings, mounting evidence now demonstrates that Wnt signaling promotes “cold tumors” formation by orchestrating an immunosuppressive TME (Fig. 3). CD103⁺ conventional type 1 dendritic cells (cDC1s) are the only cell population that mediates the transport of solid tumor antigens from the TME to the tumor-draining lymph node (tdLN), thereby activating CD8⁺ T cells [75, 76]. β-catenin signaling significantly contributes to the functional impairment of CD103⁺ Batf3-dependent cDC1s, as it induces the expression of activating transcription factor 3 (ATF3) in tumor cells, which leads to downregulation of C-C motif chemokine ligand 4 (CCL4). This impairs the activation and infiltration of CD103⁺ cDC1s into the TME. The deficiency of Batf3-dependent cDC1s results in defective priming of CD8⁺ T cells in tdLNs. Moreover, Batf3-dependent cDC1s exhibit markedly reduced expression of CXC chemokine ligands 9 and 10 (CXCL9/10), compromising their ability to engage CXCR3 on effector CD8⁺ T cells. As a result, recruitment of CD8⁺ T cells into the TME is severely impaired [74–77]. In addition, studies have shown that Wnt signaling or AKT-mediated activation of β-catenin promotes the formation of the β-catenin/TCF/LEF complex, which binds to the CD274 promoter and induces expression of programmed death-ligand 1 (PD-L1). Furthermore, β-catenin knockout leads to reduced PD-L1 expression in tumor cells, thereby enhancing CD8⁺ T cell infiltration and activation within the TME [78–82].

Immunosuppressive Foxp3⁺ regulatory T cells (Tregs) exert their suppressive functions primarily through cytotoxic T lymphocyte-associated antigen-4 (CTLA-4) and CD25. They inhibit antigen-presenting cells (APCs) via interactions with B7-H4 and indoleamine 2,3-dioxygenase (IDO). Additionally, Tregs secrete immunosuppressive cytokines such as TGF-β, IL-10, IL-4, and IL-35, which collectively impair effector cell function and disrupt their metabolic activity. Recent studies have shown that nuclear β-catenin enhances Treg infiltration, survival, and activity [8]. Targeting the interaction between B-cell lymphoma 9 (BCL9) and β-catenin suppresses Treg infiltration via the TGF-β-CCL20/CCL22 axis [83]. In melanoma, Wnt5a-induced DCs promote Treg differentiation in an IDO-dependent manner [84]. Similarly, in gastric cancer, CCL28 is a direct transcriptional target of the β-catenin/TCF complex, facilitating Treg recruitment within the TME [85].

M2-type tumor-associated macrophages (TAMs) secrete oncogenic signals that promote tumor cell migration and invasion. Notably, β-catenin, c-MYC, and AXIN2 are upregulated in M2-type TAMs. The lncRNA RP11-417E7.1 enhances expression of the adjacent gene THBS2 by maintaining specific chromatin looping and is transferred from CRCs to TAMs via exosomes, inducing M2 polarization of TAMs through activation of the Wnt/β-Catenin signaling pathway [86]. Additionally, Wnts secreted by HCC cells activate the canonical Wnt pathway in macrophages, driving their polarization into M2-type TAMs in a paracrine manner [87]. In turn, M2-type TAMs promote tumor stemness in TNBC cells by secreting VEGFA, which promotes the CSC phenotype via the Neuropilin-1 (NRP-1) receptor and the downstream GAPVD1/Wnt/β-Catenin signaling [88]. The immune checkpoint (ICP) lymphotoxin β receptor (LTβR) sustains the immunosuppressive function and M2 phenotype of TAMs through non-canonical NF-κB and Wnt/β-Catenin signaling pathways [89].

Furthermore, Wnt activation induces metabolic reprogramming from glycolysis to fatty acid oxidation (FAO) by upregulating the expression of Carnitine Palmitoyltransferase 1 A (CPT1A) upregulation, leading to DCs dysfunction and tumor progression [90]. Dysregulated Wnt/β-Catenin signaling enhances tumor cell resistance to cuproptosis and other stressors, thereby promoting tumor survival and disrupting the balance of the CI cycle [91]. Collectively, these mechanisms contribute to the establishment of an immunosuppressive TME that undermines T cell-dependent therapies.

Wnt/β-Catenin signaling pathway and tumor angiogenesis

Uncontrolled tumor cell growth results not only from genetic factors but also from abundant blood vessels that supply nutrients. Canonical Wnt signaling plays a critical role in physiological angiogenesis. For instance, Wnt2 is recognized as an angiogenic growth factor that promotes liver regeneration [92, 93]. The balance between pro- and anti-angiogenic factors maintains vascular homeostasis, and the disruption of this balance triggers the “angiogenic switch” in tumors, a critical step for tumor growth and metastasis. VEGF is a well-established downstream target of Wnt signaling, and the β-catenin/TCF complex regulates the transcription of IL-8 and MMPs, which are key mediators of angiogenesis [93]. Recent research indicates that ribosomal protein S15A (RPS15A) enhances hepatic cancer angiogenesis by enhancing Wnt/β-Catenin-induced fibroblast growth factor 18 (FGF18) expression [94]. Survivin promotes VEGF expression via a PI3K/Akt-β-catenin-TCF/LEF-dependent pathway, facilitating its release from tumor cells to drive angiogenesis [95]. In CRC, secreted WNT4 promotes disease progression by inducing EMT and activates fibroblasts to promote angiogenesis [96]. Similarly, ovarian cancer (OvCa)-secreted paracrine S100 calcium binding protein B (S100B) targets human umbilical vein endothelial cells (HUVECs) to promote angiogenesis via the FOXO1/β-catenin signaling pathway [97].

Wnt/β-Catenin signaling pathway and CAFs

Tumor cells must breach the barrier formed by the ECM, CAFs, and other stromal cells to metastasize, while proteases like MMPs downstream of Wnt signaling are critical for promoting distant metastasis [98]. Concurrently, activation of the Wnt/β-Catenin pathway within CAFs further enhances tumor progression through multiple mechanisms. Quiescent fibroblasts can be transformed into CAFs via two major pathways: first, through induction by exosomes secreted from cancer cells; second, through maintenance of the CAF phenotype by activated signaling pathways in cancer cells [9]. Once activated, CAFs contribute to immune suppression and drive tumorigenesis, therapy resistance, and metastasis [99–101]. Evidence indicates that Wnt/β-Catenin pathway activation in fibroblasts is a key driver of their transformation into CAFs [9]. Phospholamban (PLN) competitively interacts with AXIN2, thereby enhancing β-catenin activity in CAFs. In turn, this β-catenin signaling upregulates gremlin-1 (GREM1) secretion via bone morphogenetic protein 2 (BMP2) and VEGFR2, ultimately inducing EMT and angiogenesis [102]. Moreover, β-catenin activation in BRAF inhibitor-induced CAFs stimulates periostin (POSTN) secretion, contributing to drug resistance in cancer cells [103]. Additionally, CAF-derived stromal cell-derived factor 1 (SDF1) activates the CXCR4/β-catenin/PPARδ pathway, thereby promoting EMT in lung cancer [104].

In summary, beyond its intrinsic oncogenic roles, understanding how the Wnt pathway contributes to the establishment of an immunosuppressive TME is essential for developing therapeutics that target this pathway. Wnt activation promotes immune evasion through multiple mechanisms: by activating Tregs, thereby impairing APCs and T cell function; inducing M2 polarization of TAMs; suppressing CD103⁺ cDC1s via ATF3-mediated downregulation of CCL4; upregulating PD-L1; maintaining the pro-oncogenic phenotype of CAFs; enhancing cancer stemness; and promoting tumor angiogenesis. Collectively, these effects disrupt critical steps of the CI cycle, thereby accelerating tumor progression (Fig. 3).

Current status and limitations of Wnt-targeting agents

Numerous drugs targeting the canonical Wnt pathway have entered preclinical trials, primarily including antibodies and small molecule inhibitors. However, no drug targeting this pathway has yet gained approval for clinical treatment. In Table 1, we summarize the drugs currently in clinical trials. In this section, we will assess the translational potential of these agents and analyze the obstacles to drug development for the Wnt pathway.

Table 1.

Clinical agents targeting the Wnt/β-Catenin pathway

Components	Target	Cancer type	Clinical phase	Condition	Serious AEs
Vantictumab (OMP−18R5)	Fzd1/2/5/7/8	Solid Tumors	NCT01345201 (P1)	Completed (2014−05)	Diarrhea and vomiting, bone fracture with β-C-terminal telopeptide (β-CTX) increasing
		IV PC	NCT02005315 (P1)	Completed (2017−11)	Unsafe related to bone toxicity
		Non-Small Cell Lung Cancer (NSCLC)	NCT01957007 (P1)	Completed (2017−06)	Not posted
		Locally Recurrent or Metastatic Breast Cancer	NCT01973309 (P1)	Completed (2017−12)	Leukopenia and pelvic pain
OTSA101-DTPA	Fzd10	Relapsed or Refractory SS	NCT04176016 (P1)	Terminated	Not posted
Niclosamide	Fzd1	Resectable Colon Cancer	NCT02687009 (P1)	Terminated	Not posted
Ipafricept (OMP−54F28)	Wnts	HCC	NCT02069145 (P1)	Completed (2017−07)	Not posted
		Recurrent Platinum-Sensitive OvCa	NCT02092363 (P1)	Completed (2017−12)	Neutropenia and hypophosphatemia
		Solid Tumors	NCT01608867 (P1)	Completed (2017−06)	Bone fracture with β-CTX increasing
		Untreated Stage IV PC	NCT02050178 (P1)	Completed (2017−06−15)	AST elevation, nausea, maculopapular rash, vomiting and WBC decrease
WNT974 ( LGK974)	PORCN	Malignancies Dependent on Wnts	NCT01351103(P1)	Completed (2024−06−17)	No responses
		Metastatic Head and Neck Squamous Cell Carcinoma	NCT02649530 (P2)	Withdrawn	Not posted
		BRAF-mutant Metastatic CRC	NCT02278133 (P1/2)	Completed (2017−06−23)	Bone fracture
XNW7201		Advanced Solid Tumors	NCT03901950 (P1)	Completed (2022−06−30)	Not posted
RXC004/0001		Advanced Malignancies	NCT03447470 (P1)	Completed (2025−01−29)	Haematological toxicity
CGX−1321		Advanced Gastrointestinal Tumors	NCT03507998 (P1)	Unknown status	Bone resorption
			NCT02675946 (P1)
ETC−159		Advanced MSS/pMMR OvCa	NCT06513624 (P1)	Recruiting	Not posted
Rosmantuzumab (OMP−131R10)	RSPO	Advanced Relapsed Tumors Refractory Solid Tumors	NCT02482441 (P1a/b)	Completed (2018−03−28)	Not posted
E7449	TNK1/2	Advanced Solid Tumors	NCT01618136 (P1/2)	Completed (2015−07)	Fatigue, chromaturia, decreased appetite, nausea, diarrhea, constipation, and vomiting
		BRCA-Related Cancers	NCT02396433 (P1/2)	Withdrawn	Not posted
		Advanced OvCa Selected by the 2X−121 DRP®	NCT03878849 (P2)	Recruiting	Not posted
FOG−001	β-catenin/TCF	locally advanced or metastatic cancer	NCT05919264 (P1/2)	Recruiting	Not posted
ST316−101	β-catenin/BCL9	selected Advanced Unresectable and Metastatic Solid Tumors	NCT05848739 (P1/2)	Recruiting	Fatigue and ALT/AST elevation
PRI-724/ ICG 001	β-catenin/CBP	Advanced myeloid malignancies	NCT01606579 (P1/2)	Completed (2016−12−30)	Not posted
		Advanced or Metastatic Pancreatic Adenocarcinoma	NCT01764477 (P1)	Completed (2015−10)	Abdominal pain, neutropenia, and ALK phos elevation
		Advanced Solid Tumors	NCT01302405 (P1)	Completed (2015−06)	Hyperbilirubinemia
		Metastatic CRC	NCT02413853 (P2)	Withdrawn	Not posted
E7386		Advanced Neoplasms	NCT03264664 (P1)	Active, not recruiting	Not posted
		Advanced Solid Tumor Including CRC	NCT03833700 (P1)	Active, not recruiting	Nausea, vomiting, aspartate aminotransferase increased, alanine aminotransferase increased, decreased appetite, and diarrhea
		Solid Tumors	NCT05091346 (P1/2)	Completed (2024−10−15)	Not posted
		Solid Tumor	NCT04008797 (P1/2)	Recruiting	Not posted
Tegavivint (BC2059)	β-catenin/TBL1	Advanced HCC	NCT05797805 (P1/2)	Recruiting	Not posted
		Relapsed or Refractory Large B-Cell Lymphoma	NCT05755087 (P1)	Recruiting	Not posted
		Relapsed or Refractory Leukemia	NCT04874480 (P1)	Active, not recruiting	Not posted
		Metastatic EGFR-Mutant NSCLC	NCT04780568 (P1)	Recruiting	Not posted
		Recurrent or Refractory Solid Tumors, Including Lymphomas and Desmoid Tumors	NCT04851119 (P1/2)	Recruiting	Not posted

Targeting Wnt receptors or ligands

Blocking frizzled

Vantictumab (OMP-18R5) is a monoclonal antibody that specifically binds to Fzd1/2/5/7/8, competitively inhibiting the interaction between Wnts and receptors [105, 106]. In early-stage clinical trials, OMP-18R5 was combined with standard chemotherapies such as paclitaxel, demonstrating an overall response rate of 31.3% and a clinical benefit rate of 68.8% in patients with locally advanced or metastatic HER2-negative breast cancer [106]. OSTA-101 is a selective antagonist targeting Fzd10, which is administered in combination with the metal chelator diethylenetriaminepentaacetic acid (DTPA); it shows promise as a therapeutic option for synovial sarcoma (SS) [107, 108]. Niclosamide, originally developed as an anti-infective agent, exhibits notable anticancer activity primarily through inhibition of STAT3 and Fzd1 [109].

Inhibiting Wnts

Ipafricept (OMP-54F28) is a fusion protein consisting of the extracellular domain of Fzd8 linked to an IgG1 Fc fragment, functioning as a decoy receptor that sequesters Wnts [110–113]. Similar to OMP-18R5, OMP-54F28 advanced into Phase 1 trials for solid tumors, including ovarian and pancreatic cancers (PC). Clinical data demonstrated that when combined with paclitaxel and carboplatin, OMP-54F28 achieved an overall response rate of 75.7%, with complete responses observed in 29.7% of patients with recurrent platinum-sensitive OvCa. Inhibition of PORCN effectively blocks Wnt secretion, thereby suppressing Wnt signaling at its origin [114]. The first-in-class PORCN inhibitor WNT974 (LGK974) entered clinical trials in 2014 [115–117], followed by other candidates such as XNW7201 [118], RXC004/0001 [119, 120], CGX-1321 [121], and ETC-159 [122, 123]. Preclinical studies have shown that PORCN inhibitors induce regression of Wnt-dependent tumors, indicating selective targeting of pathological Wnt activation.

Inhibiting RSPO

Rosmantuzumab (OMP-131R10) is a monoclonal antibody that targets RSPO4, a secreted factor enhancing Wnt signaling by stabilizing Fzds [124]. OMP-131R10 entered Phase 1 clinical trials for advanced solid tumors. Although the maximum tolerated dose (MTD) was not reached, a target engagement-based maximum administered dose of 15 mg/kg every two weeks was established.

Targeting the DC

The DC mediates β-catenin phosphorylation and degradation, thereby maintaining pathway inactivity in the absence of Wnt signals [125]. Enhancing the activity of this complex or preventing its inhibition by Wnt signaling represents a promising therapeutic strategy for cancer. AXIN is a limiting component of the DC and is negatively regulated by tankyrase enzymes (TNKS1/2), which promote its poly-ADP-ribosylation (PARylation) and subsequent proteasomal degradation. Tankyrase inhibitors block AXIN degradation, thereby prolonging DC activity and promoting β-catenin degradation [126]. A dual PARP/tankyrase inhibitor, E7449, entered Phase 1 clinical trials in advanced solid tumors and was associated with a low incidence of toxicity [127]. To date, no selective tankyrase inhibitor has completed Phase 2 trials, underscoring the challenges of safely augmenting DC activity in patients.

Targeting β-Catenin

β-catenin, the central mediator of canonical Wnt signaling, represents a promising therapeutic target, particularly in cancers harboring APC mutations. However, direct inhibition of β-catenin has proven to be challenging; current strategies for β-catenin inhibitors primarily focus on disrupting its interactions with other components, such as the β-catenin/TCF complex [2], BCL9 [128–131], CREB-binding protein (CBP)/p300 [132], and transducin β-like protein 1 (TBL1) [133]. FOG-001 blocks the interaction between β-catenin and TCF [134]. ST316-101 is a peptide antagonist targeting the β-catenin/BCL9 interaction [135, 136]. ICG-001 (PRI-724), a selective CBP-β-catenin antagonist, specifically binds to CBP; it advanced through Phase 1 trials in advanced solid tumors [137–141]. E7386 was developed based on the microsomal stability, membrane permeability, and solubility of C-82, the active metabolite of PRI-724 [142]. Tegavivint (BC2059) exerts its anti-tumor effects by binding to TBL1 and disrupting the β-catenin/TBL1 interaction [143–145].

Limitation of the current inhibitors

Despite the advancement of several agents into early clinical trials, none have yet received regulatory approval for patient treatment. The primary reasons include (i) the inherent difficulty in targeting key components within the canonical Wnt pathway and (ii) the essential physiological roles of Wnt signaling in normal tissues, which often result in severe adverse effects (AEs) and toxicities.

Structurally, many Wnt targets lack well-defined binding pockets and display intrinsic disorder, which substantially limits the applicability of structure-based drug design strategies. For instance, the central armadillo repeat domain of β-catenin is predominantly occupied by TCF4, leaving limited space for small-molecule interference. Furthermore, β-catenin’s functionality relies on large, flat protein-protein interaction (PPI) interfaces with transcriptional coactivators, which are inherently resistant to conventional small-molecule inhibition [128, 146, 147]. Additional hurdles include β-catenin’s nuclear localization and its dynamic shuttling between cytoplasmic and nuclear compartments, complicating drug delivery and target engagement. The c-MYC protein (439 amino acids) comprises an N-terminal transactivation domain (TAD), conserved MYC homology boxes (MBs) that mediate PPIs with cofactors, and a C-terminal DNA-binding domain [148, 149]. Current therapeutic strategies against MYC focus on disrupting its heterodimerization with MAX or modulating post-translational regulators (such as SCFFBXW7, SKP2, and USP7) [150–153]. Compensatory activation of MYC paralogs (N-MYC; L-MYC) in certain cancers reduces inhibitor specificity [154]. Moreover, most candidates exhibit poor pharmacokinetic profiles, off-target effects, or insufficient potency, partly due to the intrinsically disordered TAD and lack of stable tertiary structures, which hinder structure-based drug design. In addition to these challenges, developing inhibitors for other core components (PORCN; DVL; RSPO3) faces significant challenges. Targeting PORCN is not an ideal strategy. Functionally, PORCN-mediated palmitoylation of Wnt proteins is crucial for embryonic development and adult tissue homeostasis. Additionally, the localization of PORCN on the luminal side of the endoplasmic reticulum (ER) membrane poses significant challenges for drug delivery [16]. DVL, a central cytoplasmic hub in Wnt signaling, lacks a well-defined catalytic domain, complicating drug design. Additionally, DVL interacts with pathways like planar cell polarity (PCP), potentially interfering with normal cell polarity [155–157].

While Wnt signaling is indispensable for embryonic development and tissue homeostasis, its aberrant activation drives oncogenesis and impairs immunotherapy efficacy. Most of the drugs in Table 1 were stalled or withdrawn due to severe adverse reactions. OMP-18R5, for instance, caused marked reductions in bone mineral density and pathological fractures [158–160]. Specifically, 12.5% (6/48) of patients experienced bone fractures, including a severe case requiring surgical intervention. OMP-54F28 required dose reduction due to gastrointestinal toxicity, limiting its therapeutic evaluation and resulting in limited objective response rates (ORRs). Bone-related toxicities still occurred, which required the co-administration of bone-protective agents [112]. Tankyrase inhibition has been associated with intestinal toxicity in vivo. For example, one study reported tankyrase blockade caused dose-dependent intestinal degeneration in mice, resulting in a narrow therapeutic window [161]. Clinical trials of ICG001 were halted due to grade 3 diarrhea, gastrointestinal mucosal injury, and elevated transaminase levels. Collectively, these outcomes reflect non-selective inhibition of Wnt signaling that consistently resulted in on-target toxicities.

How to overcome the “Undruggability”?

Wnt drug discovery must address both the structural complexity of its core components and their essential physiological roles. Given that dose-limiting toxicities arise mainly from effects on healthy tissues, Wnt-targeted therapies should focus on tumor-specific vulnerabilities. This approach enables selective inhibition of oncogenic Wnt signaling without compromising normal pathway functions.

Selection of appropriate targets

Blocking the Wnt pathway often harms normal tissues, causing severe AEs. Therefore, drug design should focus on tumor-specific targets to selectively disrupt Wnt signaling in tumors while sparing normal tissues. Although Wnt pathway mutations are prevalent across most cancer types, mutations in individual Wnt components display marked tissue specificity. The frequently altered Wnt pathway genes include APC, CTNNB1, AXIN1, AXIN2, ZNRF3, and RNF43 [46]. Analysis of mutation frequencies for these six genes in 33 primary cancer types listed in the MSK-CHORD (MSK, Nature 2024) database shows variable mutation patterns in CRC, Prostate Cancer (PCa), NSCLC, Breast Cancer, and PC (Fig. 4A). CRC (n = 5543) shows the highest mutation burden, with high frequencies of APC (74.67%) and CTNNB1 (6.87%) mutations. NSCLC (n = 7809) is dominated by APC (4.00%) and CTNNB1 (2.87%) mutations, while PC (n = 3109) more frequently harbors RNF43 (5.82%) mutations. Further classification of mutation types indicates that truncating and missense mutations are predominant among these six genes (Fig. 4B). Evidence suggests that APC variants truncated within the “Mutation Cluster Region” (MCR, from codon 1286 to 1581) generate the most oncogenic Wnt signals [162, 163]. Additionally, missense driver mutations in AXIN1 disrupt the structural stability of the RGS domain and promote small aggregate formation, thereby rewiring the AXIN1 interactome to promote basal Wnt pathway activation [46]. High-frequency mutations in these two genes that determine DC activity contribute significantly to cancer initiation.

Fig. 4.

Wnt pathway mutations in human cancer and expression in paired tumors and paired normal tissues. Data source: MSK-CHORD (MSK, Nature 2024) cBioPortal. A Mutation frequencies of APC, CTNNB1, AXIN1, AXIN2, RNF43, and ZNRF3 in pancancer. Top 5 frequencies are shown. B Mutation types among these six genes in pancancer. C Significant expressed Wnt/β-Catenin components in paired tumors and paired normal tissues, based on comprehensive analysis of significant expression (Log2 (TPM + 1)) across 33 tumor types in the TCGA database. Paired t tests was used to determine significant differences, **P = 0.0027, ***P < 0.001, ****P < 0.001

According to TCGA, we systematically analyzed the gene expression profiles (Log2 (TPM + 1)) of these genes across all tumor samples and paired normal tissues. The results revealed significant differences in gene expression between cancerous and normal tissues (Fig. 4C). Notably, CTNNB1 is highly expressed in thymoma (THYM) (3.92 vs. 6.18), RNF43 in colon adenocarcinoma (COAD) (1.44 vs. 6.44) and rectum adenocarcinoma (READ) (0.84 vs. 6.47), and AXIN2 in COAD (3.24 vs. 5.75) and READ (3.22 vs. 5.86). Other Wnt components also show tumor-specific expression patterns, such as Fzd7 in lung squamous cell carcinomas (LUSC) (2.75 vs. 3.95) and BCL9 in lung adenocarcinomas (LUAD) (2.57 vs. 3.65). These findings highlight that identifying tumor-specific expression profiles is a critical prerequisite for developing effective Wnt-targeted therapies. Furthermore, targeting highly mutated oncogenic drivers within the Wnt pathway represents a promising strategy for drug development. For example, although direct inhibition of β-catenin remains challenging due to its structural properties, alternative approaches focusing on disrupting its interactions with coactivators have been explored. Notably, ST316-101, a first-in-class BCL9 inhibitor designed to interfere with β-catenin/BCL9 dimerization, is currently under evaluation in Phase I/II clinical trials (NCT05848739) for advanced unresectable or metastatic solid tumors.

Diversified combined strategies

The Wnt/β-Catenin pathway interferes with cancer therapy through multiple mechanisms. As detailed in previous sections, aberrant activation of the tumor-intrinsic Wnt/β-Catenin pathway suppresses the recruitment of Batf3 cDC1s into the TME [74–76]. Combination strategies integrating tumor-targeted agents with ICIs have demonstrated enhanced antitumor efficacy, primarily by converting non-T cell-inflamed tumors into T cell-inflamed phenotypes, thereby sensitizing them to T cell-dependent killing therapies. Moreover, canonical Wnt signaling confers resistance to conventional treatments such as chemotherapy through various mechanisms, including angiogenesis, EMT, and the accumulation of somatic mutations. Consequently, a growing number of clinical trials are evaluating combinations of Wnt inhibitors with chemotherapy, ICIs, angiogenesis inhibitors, and other standard anti-tumor therapies.

Harnessing TPD technologies for targeting therapy

In recent years, emerging biotechnologies have reshaped the therapeutic landscape for previously intractable targets. Protein degradation is a fundamental cellular process that mediates protein turnover, ensures quality control in protein folding, and enables rapid responses to dynamic cellular signals. TPD strategies can be classified into two broad categories based on the mechanism of protein degradation. The first type is the ATP-dependent UPS, which primarily degrades abnormal and short-lived proteins [164]. In this process, the E1 ubiquitin-activating enzyme transfers ubiquitin molecules to the E2 ubiquitin-conjugating enzyme in an ATP-dependent manner. Subsequently, the E2 enzyme, together with an E3 ubiquitin ligase, conjugates ubiquitin to the target protein, leading to subsequent degradation by the proteasome [165]. The second category involves lysosome-dependent degradation, which primarily targets extracellular, membrane-associated, and long-lived intracellular proteins through the endosome-lysosome pathway and autophagy-lysosome pathways [164]. Compared to traditional occupancy-driven small molecule inhibitors, event-driven TPD drugs offer several advantages. TPD agents leverage their catalytic nature to directly induce the degradation of target proteins, utilizing the cellular protein degradation machinery to achieve sustained depletion of target proteins. This enables reduced dosing frequency and lower administered doses. The dual selectivity for both the target protein and the E3 ligase may reduce potential toxicity. Additionally, TPD strategies hold potential advantages in overcoming drug resistance and modulating “undruggable” targets [166–168]. For instance, a single proteolysis-targeting chimera (PROTAC) molecule can achieve sustained tumor-specific target depletion through catalytic engagement of the UPS, thereby minimizing disruption to Wnt-dependent homeostatic processes in normal tissues. Moreover, degradation-based inhibition circumvents compensatory feedback loops often associated with occupancy-driven mechanisms, significantly delaying the emergence of acquired resistance. These advantages position TPD agents as a transformative therapeutic modality for selectively suppressing oncogenic Wnt signaling while preserving essential physiological Wnt activity.

Induced targeted Wnt/β-Catenin-related proteins degradation

TPD drugs remain in the early phases of clinical development. As summarized in Table 2, several compounds that inhibit the Wnt/β-Catenin pathway through TPD strategies have been identified. This section methodically evaluates the translational potential of these compounds via a systematic review of their mechanisms of action.

Table 2.

TPD therapeutics in Wnt/β-Catenin

Name	Technologies	Target	Cancer type	Advantages	Refs
Xstax-VHL	PROTAC	β-catenin	CRC	Inhibit the survival of CRC patient-derived organoids through sustained β-catenin degradation	[172]
NP-PROTACs	PROTAC	β-catenin and STAT3	CRC	Enhance CD103+ DCs infiltration and T-cell cytotoxicity, alleviate the immunosuppressive microenvironment induced by β-catenin/STAT3	[173]
ProMyc	PROTAC	c-MYC and PD-L1		As aptamersare to solve ligand deficiency, and simultaneous degradation of c-Myc and PD-L1 to conquer tumor immune evasion	[174]
Compound 4	PROTAC	KDM3	CRC	35-fold more potency in vitro and 10-fold potency in vivo than IOX1, specially in colorectal CSCs	[175]
NRX-252114	Molecular glue	β-catenin/β-TrCP		Specificity of the enhancers for the mutant form of β-catenin without impacting the function of WT β-catenin in normal tissues, and enhancers potentiated unphosphorylated β-catenin ubiquitylation	[177]
ZNRF3*IGF1R	LYTAC	ZNRF3/RNF43	CRC	Treatment of normal and patient-derived CRC organoids led to tumour-specific degradation	[179, 180]
R2PD1	ROTAC	PD-L1	Melanoma	R2PD1 reactivates cytotoxic T cells and inhibits tumor cell proliferation more potently than Atezolizumab	[187]

UPS-dependent Wnt-TPD

UPS-dependent TPD strategies include PROTACs [169] and molecular glues [170]. PROTACs comprise three components: a protein of interest (POI)-targeting warhead, a flexible linker, and an E3-recruiting ligand [169–172]. Xstax-VHL is a PROTAC designed for degrading β-catenin. Xstax is a stapled helical peptide based on the β-catenin 469–482 amino acid binding domain of AXIN, which can penetrate the cell membrane and disrupt the interaction between β-catenin and TCF. ALAPYIP serves as a proteasome-targeting domain (PTD) and is recognized by the Von Hippel-Lindau tumor suppressor protein (VHL). Xstax binds to the VHL ligand via an amide bond using 6-aminocaproic acid as a linker to form the PROTAC peptide. Preclinical experiments have shown that Xstax-VHL effectively degrades β-catenin and can eliminate β-catenin in intestinal cancer cells [172].

NP-PROTACs are nanoengineered peptide PROTACs that efficiently degrade STAT3 and β-catenin. Peptides targeting β-catenin and STAT3 were conjugated with VHL ligands through 6-aminohexanoic acid. To enhance membrane penetration, researchers added the transmembrane peptide TAT to the end of the peptide sequence. This resulted in the formation of LAVTAT targeting β-catenin and SAVTAT targeting STAT3. LAVTAT and SAVTAT achieved optimal synergy at a molar ratio of 1:1 and were connected to DSPE-PEG2000-SH via disulfide bonds to form DSPE-PEG2000-PROTACs. The core of NP-PROTAC consists of hydrophobic lipids, the middle layer contains PEG2000, and the outermost layer is linked to the corresponding peptide PROTACs via disulfide bonds. This dual-target degradation strategy significantly enhances the killing effect of CD8⁺ T cells and inhibits M2-polarized macrophages, showing promising clinical application prospects when combined with anti-tumor immune activation [173].

c-MYC is a potential target in various cancer types and a key downstream effector of the canonical Wnt pathway. Targeting c-MYC is challenging due to its lack of a clear ligand-binding pocket. MA9C1 is an aptamer designed for c-MYC. The 5’ end of MA9C1 is modified with NH2-C6 and coupled to pomalidomide-PEG4-COOH, the ligand of the ubiquitin ligase cereblon (CRBN), to form ProMyc, which degrades c-MYC. ProMyc uses the aptamer MA9C1 to address the lack of an effective ligand for c-MYC. Additionally, researchers introduced an artificially looped anti-PD-L1 aptamer PA1 with delivery function to form circPA1-ProMyc, which simultaneously degrades PD-L1 and c-MYC, providing a novel design concept for “undruggable” targets [174].

Histone methylation and demethylation are closely related to epigenetics. Targeting histone demethylases is a research hotspot to alleviate tumor drug resistance. Histone demethylases of the KDM3 family (KDM3A, KDM3B) can epigenetically regulate the functional characteristics of stem cells in CRC via the Wnt/β-Catenin pathway. IOX1 is a histone demethylase inhibitor, and Compound 4 is composed of IOX1 as a POI-targeting warhead, PEG as a linker, and pomalidomide as an E3-recruiting ligand. Compound 4 is more effective than IOX1 alone in treating CRC both in vitro and in vivo [175].

Molecular glue has a smaller molecular weight compared to PROTACs. Molecular glue achieves targeted degradation by connecting a POI-targeting warhead with an E3-recruiting ligand. Normally, when the Wnt pathway is inactive, β-catenin is tagged for degradation by its cognate E3 ligase β-TrCP. Therefore, promoting the interaction between β-TrCP and its natural substrate β-catenin is a potential strategy for degrading β-catenin. Mutations in β-catenin phosphorylation sites, such as Ser33 and Ser37, impair the ability of β-catenin to bind efficiently to β-TrCP, leading to its stabilization and promoting an enhanced oncogenic transcriptional program [6, 176]. PPI enhancer ligands bind to the phosphoserine binding site on β-TrCP, thereby enhancing the binding of dual-phosphorylated Wnt/β-Catenin without affecting wild-type β-catenin function in normal tissues. This ensures the specificity of the enhancer for mutant β-catenin. NRX-252114 is an enhancer-based molecular glue that significantly enhances the binding of S37A mutant β-catenin to β-TrCP. The enhanced PPI affinity leads to increased K48-linked ubiquitination of mutant β-catenin via its natural ubiquitin ligase β-TrCP, thereby promoting its proteasomal degradation (Fig. 5) [177].

Fig. 5.

Preclinical TPD agents targeting the Wnt/β-Catenin pathway and Molecular mechanisms. Xstax-VHL is a β-catenin-targeting PROTAC that recruits the VHL E3 ligase complex to drive proteasomal degradation of β-catenin. LAVTAT employs a nanoengineered peptide PROTAC scaffold incorporating tandem cell-penetrating TAT motifs, enabling VHL-mediated proteasomal targeting for β-catenin depletion. NRX-252114 hijacks the endogenous β-TrCP, the canonical Wnt pathway intrinsic E3 ubiquitin ligase, to enforce β-catenin polyubiquitination and proteasomal degradation. The 5’ end of MA9C1 is modified with NH2-C6 and coupled to pomalidomide-PEG4-COOH, the ligand of the ubiquitin ligase CRBN, to form ProMyc. Compound 4: a histone demethylase-targeting IOX1 warhead connected via PEG spacer to the pomalidomide. ZNRF3*IGF1R and R2PD1 operate via a dual-targeting paradigm, direct engagement of cell-surface POI and RSPO-mediated activation of transmembrane E3 ligases RNF43/ZNRF3, ultimately routing POIs to lysosomal degradation

Lysosome-dependent Wnt-TPD

The endosome-lysosome system includes Lysosome-Targeting Chimaera (LYTAC), Bispecific Aptamer Chimera, Antibody-based PROTAC (AbTAC), and GlueTAC [15, 178]. LYTAC consists of a small molecule or antibody linked to a ligand that binds to lysosome-targeting receptors (LTRs). Bispecific Aptamer Chimera uses DNA aptamers to target proteins and transmembrane POIs to form a bispecific complex. AbTAC is a bispecific immunoglobulin that recruits transmembrane E3 ligase ZNRF3 and targets proteins simultaneously to form a ZNRF3-AbTAC-target protein complex. This complex is internalized and degrades POIs in lysosomes. Studies have shown that ZNRF3*IGF1R can achieve CRC-specific degradation, but its mechanisms, such as endocytosis and ZNRF3 recycling, require further investigation (Fig. 5) [179, 180]. GlueTAC is composed of a POI-targeting moiety, a cell-penetrating peptide (CPP), and a lysosomal sorting sequence. The POI-targeting moiety of GlueTAC promotes cell penetration using nano-antibodies instead of traditional antibodies. Additionally, covalent interactions between nano-antibodies and antigens enhance specificity and overcome off-target issues [181].

The autophagy-lysosome system encompasses several TPD technologies, including autophagy-targeting chimeras (AUTACs), autophagosome-tethering compounds (ATTECs), autophagy-targeting chimeras (AUTOTACs), and chaperone-mediated autophagy (CMA)-based degraders. AUTACs consist of a POI-targeting component, a linker, and 8-nitro-cGMP. This technology can degrade not only proteins but also organelles [182]. ATTECs are TPD agents based on the autophagy system, comprising three components: an LC3-interacting region targeting the key protein of autophagosomes, a linker, and a target protein-binding moiety [183]. AUTOTACs promote the oligomerization and activation of p62, leading to degradation via the autophagy-lysosomal pathway [184]. CMA-based degraders consist of a transmembrane domain, a POI-targeting ligand, and a CMA-targeting moiety [185].

Potential designs of TPD applications in the Wnt/β-Catenin pathway

The development of TPD drugs targeting the Wnt/β-Catenin pathway requires strategic target selection and rational degradation design, with precise engineering of three core components: the POI, the linker, and the E3 ligase.

It is essential to analyze the functional sites and mechanisms of the Wnt/β-Catenin pathway when designing Wnt-TPD drugs. RNF43/ZNRF3 can serve as the “degradation module” linked to the lysosome in LYTAC or AbTAC designs, while the POI at the other end can be selected from membrane proteins like Fzds, PD-L1, and LGR. Among these, anti-Fzd7 antibody has shown anti-tumor activity in TNBC and NSCLC [59, 186], and R2PD1 has confirmed the feasibility of degrading PD-L1 through this approach to restore T cell activity [187]. Moreover, degrading LGR can enhance the negative regulation of the Wnt pathway by RNF43/ZNRF3. Certainly, β-catenin remains the most well-validated target protein in the Wnt pathway, as half of the currently known Wnt-TPD agents target it (Table 2).

The length, chemical composition, flexibility, and attachment sites of the linker play critical roles in determining the pharmacokinetic properties of TPD drugs, including cellular permeability, metabolic stability, and solubility. Linkers are generally categorized into two types: flexible linkers and relatively rigid linkers [188]. Among these, flexible linkers are the most widely used, as they not only allow systematic adjustment of linker length but also facilitate rapid synthesis of compounds with diverse linker architectures. The most common flexible linkers include alkyl-based and PEG-based linkers. In contrast, relatively rigid linkers often feature triazole or cycloalkane motifs as key structural elements. Among the Wnt-TPD drugs summarized in Table 2, PEG-based linkers predominate. Given that β-catenin is primarily localized in the cytoplasm and can translocate into the nucleus, particular attention must be paid to cell membrane permeability during drug design. Consequently, TPD drugs targeting β-catenin often require incorporation of CPPs within the linker or resort to nanonization strategies to achieve efficient intracellular delivery.

The E3 ligase ligand is crucial for the efficacy of TPD drugs, and various E3 ligases and their ligands have been explored for cancer treatment [166]. Currently, the most commonly utilized ones include CRBN, VHL, and ligands such as pomalidomide. The Wnt/β-Catenin pathway possesses intrinsic E3 ligases within the cell, including β-TrCP and the F-box and WD repeat domain-containing protein 7 (FBXW7), which promote β-catenin degradation [189]. Therefore, leveraging β-TrCP or FBXW7 as a “degradation module” in TPD strategies offers significant advantages.

Conclusions

Wnt-TPD drugs offer novel strategies to address “undruggability”, but several challenges must be addressed before achieving clinical translation. First, while membrane targets can be efficiently degraded using AbTAC and LYTAC, the cell membrane permeability and delivery efficiency of engineered TPD macromolecules for intracellular or nuclear targets still require improvement. Another limitation lies in the selection of E3 ligases as “degradation modules”. Currently, fewer than 10 of the more than 600 E3 ligases in the human proteome have been successfully targeted by small molecules for induced protein degradation [166]. Given their highly variable mutation statuses and expression profiles across cancer types, this limited targeting capability poses a significant barrier to broad therapeutic development. For instance, RNF43-mutant or RSPO-fusion cancers frequently co-occur with FBXW7 mutations, which may confer primary resistance to therapies targeting β-catenin transcriptional activity. Although Wnt signaling drives oncogenesis in these cases, the resulting malignancy may become β-catenin-independent and instead depend on elevated c-MYC expression [189]. This poses a significant challenge for TPD drug design. Furthermore, the toxicity profiles of TPD agents compared to conventional small-molecule inhibitors remain incompletely characterized. While TPD compounds may reduce the requirement for sustained target occupancy, off-target degradation of structurally similar proteins or unintended disruption of the E3 ligase functions could result in unforeseen AEs. In addition, potential risks associated with TPD agents include toxicity arising from cleaved linker fragments, excessive target protein degradation, and the formation of potentially toxic binary complexes such as PROTAC-POI and PROTAC-E3 ligase adducts [166, 190, 191].

Most existing Wnt-targeted clinical agents have failed to progress further due to severe AEs (Table 1). This is primarily because the Wnt pathway plays a critical physiological role in maintaining normal tissue homeostasis. However, accumulating evidence indicates that aberrant Wnt activation also contributes to the low response rates of current ICIs. As a result, therapeutic strategies for Wnt-driven tumors have shifted from direct pathway inhibition toward combination with immunotherapy. Given that Wnt activation directly upregulates PD-L1 expression, precise modulation of the Wnt pathway is expected to effectively reverse the non-T cell-inflamed TME. For example, R2PD1, a TPD drug based on RNF43/ZNRF3, is a chimeric adhesin that fuses the RSPO2 Furin domain with PD-1 to induce PD-L1 degradation (Fig. 5). As a monotherapy, it demonstrates superior efficacy to Atezolizumab [187]. Moreover, preclinical studies on NP-PROTACs have shown enhanced infiltration of CD103⁺ cDC1s and increased T-cell cytotoxicity. Several clinical trials evaluating the combination of Wnt inhibitors with PD-1/PD-L1 inhibitors have reported promising results (NCT05091346, NCT04907539, NCT02013154). Notably, Fzd7 exhibits endocytic activity [192], and anti-Fzd7 antibodies have demonstrated potent antitumor effects. Collectively, these findings suggest that a bispecific degrader targeting both Fzd7 and PD-L1 could potentially block the Wnt pathway while simultaneously restoring T cell infiltration in the TME.

The immunosuppressive TME induced by Wnt activation is a major contributor to ICIs treatment failure. Targeting the Wnt pathway alone with TPD degraders is insufficient to reverse the immunosuppressive TME. In particular, macromolecular agents such as PROTACs often require incorporation of CPPs into their linkers or nanoformulation of TPD molecules to achieve effective intracellular delivery and target protein degradation. Therefore, a key determinant of Wnt-TPD drug efficacy is the ability to achieve targeted delivery of TPD degraders while simultaneously modulating the TME. Tumor cells with activated Wnt signaling can release exosomes that regulate the functions of immunosuppressive cells such as CAFs and macrophages. Moreover, CAFs not only impede the penetration of exogenous drugs into tumor tissues but also express PD-L1, forming “immune traps” [193]. Studies have shown that encapsulating the Wnt inhibitor ICG-001 in nanoliposomes and combining it with an anti-PD-L1 antibody can reprogram CAFs into quiescent fibroblasts, thereby enhancing the anti-tumor immune response [9]. This strategy offers valuable insights for the development of Wnt-TPD degraders based on nanodelivery systems.

Overall, this review provides a systematic analysis of the central role of the Wnt/β-Catenin pathway in oncogenesis and immune regulation, offering novel insights into addressing the “undruggable” problem of this pathway. It comprehensively discusses the design principles and recent advances in Wnt-TPD drugs. To address the challenge that core proteins such as β-catenin are structurally difficult to target directly, Xstax-VHL integrates a stapled helical peptide with TPD technology to address both the challenge of targeting β-catenin and enhancing peptide drug penetration, thereby effectively blocking the β-catenin/TCF interaction. Similarly, ProMyc utilizes the aptamer MA9C1 to overcome the lack of effective binding sites for c-MYC. Moreover, frequent mutations in genes such as APC, CTNNB1, AXIN1, AXIN2, RNF43, and ZNRF3 contribute to various cancers [46]; the design of mutation-specific therapeutics may help overcome tumor resistance. A notable example is NRX-252114, which exhibits enhanced specificity for mutant β-catenin without affecting the function of wild-type (WT) β-catenin in normal tissues, significantly reducing the risk of toxicity associated with targeted therapy. Importantly, recent studies have shown that TPD-based bispecific antibody (BsAb) designs enable logic-gated extracellular protein degradation with cell-specific selectivity [194], supporting the potential of Wnt-TPD strategies for precise and conditional target degradation. In conclusion, although drug development targeting the Wnt pathway still faces multiple challenges, these innovative approaches aimed at reducing toxicity and enhancing the response to ICIs are poised to provide critical support in overcoming therapeutic barriers in Wnt-driven cancers.

Author contributions

SM conceptualized the review topic, and drafted the original text. XZ and YZ conducted literature search and revised the manuscript. XL and ZW drew the figures and listed the table. PZ and ZW completed the partial revisions. XZ interpreted the adverse reactions of the Wnt related inhibitors from a clinical perspective. XH supervised the review writing. GH and WX is guarantor of the review and contributed to the critical discussion. All authors reviewed and approved the final manuscript.

Funding

This review was supported by the Natural Science Foundation of Shanghai (23ZR1427400); “AI Empowerment for Research Program” initiated by the Shanghai Municipal Education Commission (SHJWAIJK241205); Key Clinical Program of Shanghai Municipal Health Commission (20214Y0516); Construction project of Shanghai Key Laboratory of Molecular Imaging (18DZ2260400); National Natural Science Foundation of China (82127807) and National Natural Science Foundation of China (82203714).

Data availability

The datasets generated and/or analysed during the current study are available in the MSK-CHORD (MSK, Nature 2024) and TCGA.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.