Mapping Cellular Stress Biology to Tackle Undruggable Targets

Episode Summary

Executive Summary: Soleil Therapeutics’ CEO Yaram Gazarian explains a science-first drug discovery platform that uses AI and live-cell imaging to map how cells sense and respond to stress, then identifies phenotype-shifting compounds and works backward to mechanism and indication. The company has already generated multiple oncology programs, with two lead small molecules advancing toward the clinic and broader applications in neurodegeneration, metabolic disease, and regenerative indications.

Main Topics: Stress biology as the basis of drug discovery (Priority: 5/5): Gazarian argues that cell stress is central to disease and that drugs should be discovered by observing how they alter cellular stress responses rather than by starting with a single target. AI-enabled imaging platform and data generation (Priority: 5/5): Soleil uses multimodal live-cell imaging and AI to capture thousands of intracellular features over dose and time, building models from large compound screens and cellular phenotypes. Flipping the traditional drug discovery paradigm (Priority: 5/5): The company starts with phenotype and cellular response, then deconvolutes mechanism and selects disease/patient populations later, claiming this overcomes a key weakness of phenotypic screening. Lead oncology programs and clinical strategy (Priority: 5/5): SOL001 is being advanced in AML and SOL002 in solid tumors, each with distinct scaffold and mechanism, with IND-enabling work underway and clinical trials planned. Expansion beyond oncology (Priority: 4/5): The same stress-biology framework is being applied to Parkinson’s, obesity, hair loss, stroke, and cardiovascular disease, with some preclinical validation already published. Capital efficiency, execution, and partnering (Priority: 4/5): Gazarian emphasizes that the company has created a broad pipeline with a small team and substantial capital efficiency, while remaining open to partnerships on select programs and platform opportunities.

Key Arguments: Cell stress is a common driver across many diseases, so measuring stress responses can reveal drugs that either increase stress in cancer cells or reduce stress in non-oncology settings. Traditional target-first discovery works best in simple monogenic diseases, but complex diseases require a broader, phenotype-based approach. Soleil’s platform measures thousands of image-based cellular features, letting the company identify a drug’s global cellular effects rather than only its presumed target. The company can deconvolute mechanism of action after discovering a phenotype-shifting compound, addressing a longstanding challenge in phenotypic screening. Soleil’s programs are selected based on both scientific merit and practical development considerations, including unmet need and clinical tractability. The company claims strong preclinical selectivity and tolerability, including activity in cancer cells with non-cytotoxicity in normal human cells. The platform’s data can support target deconvolution, indication selection, and eventually prediction of efficacy and toxicity. Soleil is not positioning itself as AI-first; AI and automation are tools that scale a long-running biology program. The same core biological logic extends to neurodegeneration, metabolic disease, and regeneration because many conditions reflect progressive cellular stress and loss of function. The company has been able to raise capital because it shows working data, not just a conceptual platform, and because it has advanced assets into IND-enabling studies. Partnerships are likely to be selective and strategic, with some core programs retained to build enterprise value and others potentially partnered to accelerate development. The company claims a small team and focused budget have still produced more than 10 novel oncology programs, underscoring capital efficiency.

Data Points: Company age: 5 years - Gazarian notes the company itself is five years old, though the science dates back over 16 years. Collaborative research timeline: 16+ years - He says he and his co-founder have been working on the platform for more than 16 years. Compound library size: close to 500,000 small molecules - He says Soleil now has a growing compound library of nearly half a million molecules. Screening throughput: over 100,000 drugs per week - The company says its wet lab and automation systems can screen this volume weekly. Novel oncology programs: more than 10 - Gazarian says the platform has generated more than 10 novel oncology programs in under two years. Lead candidates: 2 first-in-class experimental small molecules - Two programs are being advanced toward the clinic. Funding raised: $290 million - Gazarian cites total capital raised since founding. Recent financing: $200 million Series C - He identifies a recent Series C as part of the total raised. Team size: less than 40 people - He emphasizes capital efficiency and a small operating team. Target milestone: IND submission by end of year - For SOL001, the company plans to file an IND by year-end if enabling work succeeds. Clinical timing: by 2027 - Gazarian expects two concurrent trials by 2027. Disease focus for SOL002: head and neck squamous cell cancer - He says this is the likely initial indication for the solid tumor program. Disease focus for SOL001: acute myeloid leukemia (AML) - The first lead oncology program is being advanced in AML. Prior validation areas: 2 published indications - He says peer-reviewed validation exists in cardiovascular and stroke settings. Platform foundation investment: $7–8 million - He says he raised this amount in philanthropy to validate the science before founding the company.

Pivotal Quotes: "Instead of starting with disease and target and drug, let's find drugs that give rise to these unique Soleil phenotypes." — Yaram Gazarian: He explains the company’s inversion of the conventional target-first discovery model. "We are not an AI-first company. If anything, we're a science-first company." — Yaram Gazarian: He distinguishes Soleil’s biology-driven thesis from many AI drug discovery startups. "You want to have selectivity, frankly, within tumors, but also against non-tumor cell lines to have a therapeutic window." — Yaram Gazarian: He describes the desired profile of the oncology programs and why stress selectivity matters.

Implications: Soleil’s approach suggests AI-enabled phenotypic discovery can produce differentiated assets for both cancer and non-cancer diseases, potentially improving hit quality, target deconvolution, and capital efficiency. If clinical data validate the platform, it could reshape early drug discovery and partnering models.