The focus of cancer drug development has been shifting in recent years. Unlike the previous emphasis on optimizing established targets, the latest wave of innovation is concentrated on breaking through targets long considered "undruggable." At the 2025 European Society for Medical Oncology (ESMO) Targeted Anticancer Therapies (TAT) congress, results from multiple first-in-human studies for innovative therapies targeting such proteins were presented, indicating the field is accelerating from basic research to clinical validation. Subsequently, Annals of Oncology systematically reviewed representative studies, highlighting drug candidates that have demonstrated mechanism validation in humans, shown biomarker-related signals, and exhibited manageable safety profiles. The review also covered several innovative technology platforms still in preclinical stages but showing potential to overcome "undruggable" limitations. This article summarizes the early breakthroughs and their potential significance in the "undruggable" target space, outlining the latest developments in this cutting-edge area.
Target: MYC MYC has long been considered a classic "undruggable" target. Its protein structure is highly dynamic and lacks stable binding pockets, making it difficult for traditional small molecules to achieve effective binding and selective modulation. Early research attempted to indirectly downregulate MYC expression using BET inhibitors, but this approach faced challenges in the clinic due to toxicity and insufficient efficacy. With advances in protein engineering and chemical biology, intervention strategies for MYC have diverged into two main paths: direct functional blockade and protein degradation. In the functional blockade approach, the mini-protein therapy OMO-103 inhibits MYC–MAX dimerization, directly interfering with MYC-driven transcription programs. In its first-in-human study, the drug showed encouraging early signals of activity with a manageable safety profile; adverse events were primarily Grade 1–2, without the dose-limiting toxicities commonly associated with earlier BET inhibitors. The drug was observed to effectively enter tumor tissue and produce early clinical activity in various advanced solid tumors, mainly manifesting as disease stabilization in colorectal, pancreatic, and other gastrointestinal cancers, with most patients having received multiple prior lines of therapy. Although these results are early signals, they represent the first validation in human studies of the feasibility of directly modulating MYC function. A phase 2 study of this therapy for advanced high-grade osteosarcoma is now underway. Concurrently, in the protein degradation field, MYC-targeting proteolysis-targeting chimeras (PROTACs) designed to recruit the VHL E3 ligase can induce dose-dependent clearance of endogenous MYC protein via ubiquitination, accompanied by decreased expression of MYC-related genes and inhibition of tumor cell proliferation. In models of MYC-amplified breast and prostate cancer, these molecules demonstrated clear biological activity. Researchers are further optimizing the structure towards more drug-like small molecules, indicating potential for future development as oral degraders or molecular glues.
Target: KRAS KRAS is one of the most emblematic "undruggable" targets in oncology, with drug development challenges stemming from its unique molecular biology. The KRAS protein has extremely high affinity for GTP/GDP, while its surface lacks stable small-molecule binding pockets, long limiting traditional inhibition strategies. Advances in structural biology and covalent chemistry led to the initial breakthrough with KRAS G12C inhibitors, with sotorasib and adagrasib subsequently receiving FDA approval. KRAS targeting strategies are thus evolving from single mutation inhibition towards broader precision intervention. In early clinical studies, the RAS(ON) tri-complex inhibitor zoldonrasib (RMC-9805) represents a significant advance. At the recommended phase 2 dose, the objective response rate reached 61% with a disease control rate of 89% in previously treated non-small cell lung cancer patients, with no dose-limiting toxicities observed, suggesting clear pharmacological control of the KRAS G12D mutation subtype in humans. Further studies showed early activity in a cohort of pancreatic cancer patients with KRAS G12D mutations, with an objective response rate of 30% among 40 patients, alongside significant decreases in mutant allele frequency in circulating tumor DNA, reaching undetectable levels in some patients. Meanwhile, KRAS degraders are also showing potential for cross-mutation type intervention in preclinical and early translational research. HDB-82, a selective protein degrader for KRAS G12D, achieved over 90% KRAS protein degradation in vitro, concurrently inhibiting MAPK and PI3K/AKT signaling pathways, and demonstrated significant tumor growth inhibition in xenograft models. Another candidate, RP04340, offers oral bioavailability and can degrade KRAS G12C, G12D, and G12V mutant proteins, achieving over 80% protein degradation and effectively inhibiting downstream MAPK signaling in various models, further supporting the potential of protein degradation strategies to overcome KRAS resistance and mutational heterogeneity.
Target: MTAP A more challenging strategy than directly inhibiting oncogenic proteins involves exploiting tumor-specific metabolic vulnerabilities for selective killing, with MTAP deletion being a prime example. While MTAP gene deletion is common in many cancers, it is not a traditional oncogenic driver mutation, and its drug development has long lacked a clear path. Research found that MTAP-deficient cells develop a specific dependency on the PRMT5–MAT2A signaling pathway, creating a new therapeutic window for synthetic lethality strategies. In early clinical studies, the SAM-competitive PRMT5 inhibitor AMG193 leverages the accumulation of methylthioadenosine (MTA) in MTAP-deficient tumors to achieve selective inhibition of cancer cells, with preliminary anti-tumor activity observed in patients with various advanced solid tumors. The study showed objective responses and disease stabilization signals in heavily pre-treated patients, while drug exposure levels correlated with changes in target inhibition biomarkers, indicating mechanism validation in humans. Overall safety was manageable, though some patients experienced dose-related adverse events, suggesting researchers are still exploring the optimal therapeutic window. Concurrently, candidates like MRTX1719 and TNG462 are advancing in early clinical studies, further validating the broad applicability of the MTAP deletion synthetic lethality strategy. Preliminary results show anti-tumor activity trends across different tumor types, including tumor shrinkage or sustained disease control signals, suggesting efficacy may depend more on molecular subtype than tumor origin. Dose-limiting toxicities such as nausea and electrolyte disturbances observed in some study cohorts indicate this new mechanism still requires optimization to balance selectivity and tolerability.
Target: TP53 TP53 is one of the most frequently mutated genes in human cancer. Unlike typical oncogenes that are abnormally activated, TP53 mutations often result in loss of tumor suppressor function, making traditional inhibitor strategies ineffective. Therefore, restoring, rebuilding, or substituting p53 function has been a central challenge in oncology drug development for decades. For tumors retaining wild-type TP53, MDM2 inhibitors can restore p53 function by relieving its inhibition; early studies confirmed this mechanism activates the p53 pathway and inhibits tumor growth, but hematological toxicity has limited the therapeutic window for such therapies. New-generation MDM2–p53 antagonists are now aiming to improve tolerability through optimized pharmacokinetics and selectivity design. For mutant TP53, structure-driven small molecule correction strategies represent a significant breakthrough direction. PC14586, which targets the TP53 Y220C mutant, has shown preliminary clinical response signals in patients with ovarian, breast, and endometrial cancers in early clinical studies, accompanied by molecular evidence of p53 function restoration. APR-246 (eprenetapopt) promotes refolding of mutant p53 and modulates cellular redox status, partially restoring tumor suppressor function, and has shown biological activity and disease control signals in early studies.
Target: WNT/β-catenin Pathway The WNT/β-catenin pathway represents a complex signaling network that is physiologically essential but co-opted by tumors, posing inherent safety challenges for therapeutic development. This pathway plays a critical role in embryonic development and tissue homeostasis, and early inhibition strategies were often limited by mechanism-based adverse effects such as bone toxicity. Consequently, achieving selective regulation of tumors while preserving normal physiological function has become a central challenge for WNT-targeted therapy development. Clinical results show that the PORCN inhibitor zamaporvint (RXC004), targeting an upstream key enzyme in the WNT pathway, combined with the PD-1 inhibitor nivolumab, provided sustained clinical benefit in some patients with microsatellite stable colorectal cancer harboring RNF43 or RSPO abnormalities, achieving a disease control rate of 57.1%. The study also observed molecular response signals with radiographic changes correlating with decreases in circulating tumor DNA (ctDNA), suggesting that upstream modulation of WNT signaling may re-establish sensitivity to immunotherapy in specific molecular subgroups. Beyond upstream inhibition, downstream functional modulation strategies are also advancing. The β-catenin–TBL1 complex inhibitor tegavivint, which interferes with β-catenin-mediated transcriptional regulation, demonstrated good tolerability and preliminary anti-tumor activity signals in early studies. Meanwhile, alternative targets like GPC3 are emerging as new entry points for indirectly intervening in WNT signaling. The bispecific antibody SAR444200 recruits immune effector cells for targeted attack on tumors, showing T-cell activation and anti-tumor activity trends in early studies. Overall, these "bypass" intervention models suggest that for structurally complex, physiologically essential signaling networks, achieving functional modulation through downstream nodes or alternative pathways may be a crucial route to enhancing therapeutic feasibility and safety windows.
The history of cancer treatment continually demonstrates that breakthroughs that truly change clinical practice often originate from seemingly faint scientific signals in early stages. From KRAS to MYC, and from TP53 to the WNT pathway, targets once deemed undruggable are now progressively entering clinical development through various technological pathways, including protein degradation, synthetic lethality, immune redirection, and engineered biologics.
Comments