UK-based Charm Therapeutics has closed an $80 million Series B round as it seeks to establish a foothold in a rapidly expanding and increasingly competitive menin inhibitor market. This class of therapies is being developed for acute myeloid leukemia (AML), a space shaped in recent months by commercial breakthroughs from Syndax and Kura Oncology.
This latest funding round was co-led by New Enterprise Associates (NEA) and SR One, with additional participation from NVIDIA, and is intended to advance the company’s lead asset into clinical development in early 2026.
Despite entering the clinic later than its rivals, Charm has assembled an investor syndicate that includes OrbiMed, F-Prime, and Khosla Ventures, bringing its total capital raised to more than $150 million. The core investment thesis centers on a distinct technical approach. While most structure-based drug discovery platforms rely on static docking models, Charm’s proprietary DragonFold system uses AI to predict the co-folding of both the protein and ligand in a dynamic environment.
The company reports that this method has produced a molecule capable of retaining activity across all currently known clinical resistance mutations in KMT2A-rearranged and NPM1-mutant acute myeloid leukemia. Should these early findings translate into the clinic, Charm could secure a meaningful competitive edge in a field now defined as much by durability and resistance management as by initial efficacy.
Menin inhibitor commercial landscape
Charm Therapeutics is entering a market that has evolved quickly over the past year, as menin inhibitors have transitioned from an emerging therapeutic concept to a commercially competitive drug class.
Syndax’s Revuforj (revumenib), approved in November 2024 for KMT2A-rearranged AML and in October 2025 for NPM1-mutant disease, maintains first-mover advantage and posted $32 million in revenues in Q3 2025. However, its long-term durability remains uncertain as resistance mutations continue to surface, prompting further clinical evaluation and mitigation strategies.
Kura Oncology’s Ziftomenib (Komzifti), approved in November 2025 for relapsed or refractory NPM1-mutated AML, has followed as a second entrant with a more favorable safety profile compared to earlier-generation agents.
With this, Charm is positioning its candidate, CHM-029, as a next-generation therapy engineered to overcome the resistance and durability limitations emerging in first-wave commercial products. The company expects to begin first-in-human testing in Q2 2026.
Foundational IP for DragonFold’s deep-learning system
A significant part of Charm Therapeutics’ valuation rests on its IP portfolio, now strengthened by the issuance of U.S. Patent No. 12,211,588. The patent protects the core computational framework behind DragonFold, the company’s AI-driven discovery platform. At its center is a deep learning method that predicts how a protein and drug candidate fold together in three dimensions, using only the protein’s amino acid sequence and the ligand’s molecular structure. This positions DragonFold as a system for true protein–ligand co-folding rather than conventional static structure prediction.
While AlphaFold transformed the field by predicting individual protein structures, Charm’s patent extends the concept to binding interactions, a capability critical for rational small-molecule design. Traditional docking tools typically assume proteins are rigid, despite evidence that they change shape as molecules bind. DragonFold’s patented model instead accounts for movement on both sides of the interaction, training a neural network to refine predicted poses based on rotational, translational, and side-chain errors.
The accompanying diagram outlines how the platform operates: stored structural datasets, training loss functions, and atomic coordinates feed into a processing system that prepares inputs for dedicated protein and ligand embedders. Their outputs are integrated through a 3D-invariant, bond-aware transformer that models how both components fold in relation to one another. The result is a dynamic, interaction-aware prediction engine designed to generate more realistic drug–target structures than traditional one-directional approaches.
The patent, titled “Methods and systems for using deep learning to predict protein-ligand complex structures”, was filed on October 2, 2023, and published on January 28, 2025. The patent, currently assigned to Charm Therapeutics Limited, lists Lakshyaditya Singh Aithani and Liam Atkinson as inventors. Cooley LLP represented Charm.
Menin Inhibitor: Patenting Activity
The patenting activity for menin inhibitors in AML shows an early discovery surge, driven by foundational works and important preclinical validation, including Kura Oncology’s study demonstrating that their menin-MLL inhibitor effectively suppressed leukemic progression. This confirmation helped trigger a spike in granted patents in 2016, as multiple groups moved aggressively to secure foundational IP around menin-KMT2A inhibition.
From 2017-2020, filing activity stabilized as companies shifted toward refining chemical series, formulations, and therapeutic strategies. Notable examples include Janssen’s 2017 optimization of its exo-aza spiro scaffold that helped mature the chemistry behind its investigational oral drug (WO2019120209A1, Biomea’s 2018 covalent inhibitor filing that fine-tuned electrophile warheads and binding motifs for what would become BMF-219 (US11084825B2), and Syndax’s 2020 patent on CYP3A4-boosted dosing strategies to enhance menin inhibitor efficacy (US20230165858A1).
The space then enters a breakout phase in 2021, coinciding with expanding clinical programs and growing confidence in menin inhibition as a viable AML therapeutic modality. The filing trend in the following years reflect sustained competition and ongoing efforts to develop next-generation molecules addressing resistance, safety, and differentiation claims. For Charm Therapeutics, this crowded but evolving landscape creates a strategic opening: with first-generation agents now revealing resistance liabilities and safety/tolerability ceilings, Charm’s DragonFold-enabled design platform allows it to enter the field not as a late follower, but as a potential best-in-class challenger engineered precisely to overcome the shortcomings highlighted by a decade of prior patents and clinical experience.
Menin Inhibitor: Top Assignees
The patent landscape reveals a diverse mix of large pharmaceutical companies, emerging biotechs, and academic institutions shaping innovation around menin inhibition and related AML technologies. Janssen Pharmaceuticals holds the most expansive portfolio, reflecting its deep medicinal chemistry investment and the extensive scaffold optimization behind bleximenib. Kura Oncology follows as a major filer, and its strong presence is closely linked to the University of Michigan, which also stands out as one of the most active academic entities in this space.
Among biotechs, Vitae Pharmaceuticals, a subsidiary of Allergan and Biomea Fusion represent two significant contributors. Syndax Pharmaceuticals maintains a sizable IP portfolio (with 54 applications) focused on revumenib, including dosing, formulation, and combination strategies. Additional contributions from Children’s Hospital Medical Center, Sumitomo Pharma, HHS, and Kurome Therapeutics illustrate the breadth of both academic and industry engagement and collaboration in this drug class.
Menin Inhibitor: Top Technology Areas
The CPC distribution underscores that menin inhibitor innovation has been overwhelmingly anchored in therapeutic intent (A61P) and drug preparation/medicinal formulation (A61K). The dominance of these two categories reflects the field’s primary objective: developing clinically viable small-molecule therapies for AML, particularly KMT2A-rearranged and NPM1-mutant subtypes.
The third major category, C07D (heterocyclic compounds), tracks closely with these therapeutic codes, highlighting the chemistry-driven nature of the space. The presence of C07B and C07F/C07C further indicates active exploration of synthetic routes, nonstandard heterocycles, and alternative core architectures during optimization cycles. The CPC profile paints a landscape heavily centered on medicinal chemistry innovation, scaffold refinement, and clinical translation, with supporting biological and analytical technologies emerging later.
For Charm Therapeutics, this pattern is particularly advantageous: the concentration of prior patents in conventional heterocycles (C07D) and traditional medicinal chemistry categories (A61K/A61P) highlights a clear gap for structurally novel, AI-designed scaffolds, an area where Charm’s DragonFold platform can generate differentiated chemical space that avoids prior art while directly addressing resistance, potency, and toxicity limitations seen in first-generation menin inhibitors.
Menin Inhibitor: Top Legal Representatives
The distribution of legal representatives in the menin inhibitor patent space reflects both the geographic concentration of filings and the strategic priorities of major pharmaceutical players. China Patent Agent (H.K.) Ltd. leads the list, supported by individual attorneys like Peng Chang and Chu Mingming. Their prominence mirrors China’s position as one of the top jurisdictions for menin inhibitor filings and highlights the role of specialized Chinese IP firms in managing high-volume chemistry and therapeutic patents
Global firms such as Gowling WLG (Canada) and Spruson & Ferguson (Australia) also rank strongly reflecting active filing pipelines across Canada and the Asia-Pacific region. The presence of Johnson & Johnson’s Patent Law Department through key internal counsel Philip Lenaerts aligns with Janssen’s deep medicinal chemistry work on menin inhibition. A second tier of individual Japanese and regional practitioners through Hiroshi Kobayashi and Norio Ohmori, underscore Japan’s significant presence as both a filing jurisdiction and a research hub for small-molecule oncology programs
The concentration of leading representatives in China and the presence of large in-house teams such as Johnson & Johnson’s reflect how heavily resourced incumbents protect their menin inhibitor portfolios across major global markets. Charm Therapeutics’ recently granted patent (U.S. Patent No. 12,211,588) is represented by Cooley LLP, one of the top firms appearing in the dataset. This aligns Charm with a high-tier U.S. legal team experienced in life-sciences IP and signals that the company is building its patent strategy with the same sophistication as established pharmaceutical players, an important foundation as it enters a space crowded with well-defended first-generation menin inhibitor portfolios.
Menin Inhibitor: Top Jurisdictions
The jurisdictional distribution of menin inhibitor patents reflects a globally competitive and commercially oriented protection strategy. The United States leads with 158 filings, underscoring its role as the largest single pharmaceutical market and the jurisdiction where early clinical programs (revumenib, ziftomenib, BMF-219) have been most active. China and the European Patent Office closely follow, demonstrating strong interest in securing rights across major global markets. The presence of other jurisdictions represent strategically important secondary markets where innovators typically extend their coverage once primary US-EU-China filings are secured.
Overall, the filing pattern shows that companies developing menin inhibitors view this modality as a global commercial opportunity rather than a regionally niche therapeutic. For Charm Therapeutics, this means that any best-in-class challenger must be IP-positioned not only in the U.S. and Europe but across the same multinational footprint.
Next-wave opportunities
Looking ahead, Charm Therapeutics enters the next phase of the menin inhibitor race with strong momentum: fresh capital, a differentiated AI-driven design engine, and a clear opportunity created by emerging resistance in first-generation drugs. If CHM-029 delivers clinically, Charm could redefine the competitive landscape and set a new benchmark for next-wave AML therapeutics.
