Finding New Targets on the Surface of Misfolded Proteins

Episode Summary

Executive Summary: The episode features Emuto Scientific co-founder and CEO Feresh Choudhry explaining how the company uses AI-enabled structural surfonomics to identify disease-specific protein conformations, especially in cancer, to improve target selectivity and reduce toxicity. The approach aims to open new drug targets, rescue failed programs, and support an internal pipeline led by an AML ADC, while also building partnerships and expanding into immunology and inflammation.

Main Topics: Drug toxicity as a central bottleneck in oncology (Priority: 5/5): Choudhry frames toxicity as a major cause of clinical failure and a limiting factor in cancer treatment, driven by shared targets across healthy and diseased tissues. Disease-specific protein conformations as a new target class (Priority: 5/5): The company’s core thesis is that proteins adopt distinct disease conformations in diseased cells, creating unique epitopes and binding sites that can be selectively targeted. Structural surfonomics platform and AI-driven target discovery (Priority: 5/5): Emuto combines structural biology, proteomics, and AI to map conformational changes at omics scale, prioritizing targets and designing binders against disease-specific epitopes. Pipeline workflow from discovery to therapeutic design (Priority: 4/5): The conversation walks through the process: patient-derived models, target ranking with an internal LLM, antibody design, validation in patient samples, and modality selection such as ADCs or bispecifics. Lead AML program IMTO-4842 and rationale for focusing on AML (Priority: 5/5): The company’s lead asset is an AML ADC designed to hit a disease-specific conformation while sparing healthy hematopoietic cells; AML is chosen for its unmet need, toxicity burden, and clear regulatory path. Business model: hybrid platform revenue plus internal therapeutics (Priority: 4/5): Emuto evolved from CRO-like analytical services into a hybrid model that includes pharma partnerships, data licensing, co-development, and internally owned drug programs. Expansion beyond oncology into immunology, inflammation, and other diseases (Priority: 3/5): The platform is positioned as broadly applicable to any disease involving structural dysregulation, with planned expansion into immunology/inflammation and potential relevance to neurodegeneration.

Key Arguments: Toxicity is a primary reason drugs fail, and in oncology the problem is amplified because many targets are shared with healthy tissue. Sequence-level targeting is insufficient; protein conformation can differ between healthy and diseased states and can be used to distinguish them. Structural biology and standard proteomics each miss critical information: one is low-throughput and static, the other high-throughput but non-structural. AI is necessary because the platform generates rich, multidimensional structural proteomics data that must be triaged and ranked to find actionable targets. Disease-specific epitopes can enable very high selectivity, with the company claiming selectivity ratios of 1,000:1 or greater. A major value proposition is rescuing failed programs that may have hit the wrong epitope rather than being fundamentally invalid. The platform may also expose binding pockets in proteins previously considered undruggable because disease conformations can reveal new accessible sites. AML is a strong initial focus because it has high unmet need, limited durable options after relapse, and a relatively established regulatory path for novel mechanisms. The company’s hybrid model is designed to fund platform development, validate the science with pharma partners, and build long-term value through internal assets. Expansion into immunology and inflammation will prioritize diseases where structural changes under stress biology are likely and commercially meaningful.

Data Points: Clinical failures due to toxicity: About one-third - Choudhry said roughly one-third of all clinical failures are due to toxicity and safety concerns. Therapeutic index crowding: 9 assets per cancer target - He cited a McKinsey study indicating an average of nine assets are being developed for each cancer target. Target selectivity: 1,000:1 or greater - Emuto claims this level of selectivity for some disease-specific binders. Academic medical center partners: 3 - The company said it works with the University of Wisconsin, University of Pennsylvania, and Mass General Hospital. Sample cohort size for validation: 50 to 100 samples - Used to screen prevalence of targets in patient cohorts and compare against healthy donors. Pharma partners worked with: 8 of the top 10 pharma companies globally - Choudhry described the company’s prior analytical partnership base. Seed financing: $8 million - The company recently closed an oversubscribed seed round originally targeted at $5 million. Initial seed target: $5 million - The round was expanded due to investor demand. Clinical burden of AML: Over 100,000 patients per year - He cited annual AML diagnoses as part of the rationale for focusing on the disease. Timeline for IND: 2027 - The company plans to establish its regulatory team to obtain an IND in 2027. Timeline for clinical trials: End of 2027 - The company expects to initiate clinical trials toward the end of 2027. Future Series A timing: Next year - Choudhry said the firm aims to raise a Series A next year after completing key in vivo and NHP studies. Lead program type: ADC - IMTO-4842 is described as an antibody-drug conjugate for AML.

Pivotal Quotes: "About one-third of all clinical failures are actually due to toxicity and safety concerns." — Feresh Choudhry: Explaining why toxicity is a central problem in drug development and oncology. "Structure provides a pathological signature of the disease, which now we can see for the first time with the platforms that we have developed." — Feresh Choudhry: Describing the company’s core scientific insight behind structural surfonomics. "We see selectivity ratios of thousand to one or greater." — Feresh Choudhry: Describing the degree of specificity achieved by the platform’s therapeutics.

Implications: If validated clinically, Emuto’s approach could widen therapeutic windows, revive shelved drug ideas, and open targets beyond sequence-based drug discovery. It may also shift partnering economics toward platform-enabled precision targeting.