Pathos

What Pathos does

Pathos is a clinical-stage, AI-enabled precision oncology biotech building a new development engine for precision oncology through multimodal data, biological modeling, and biopharma partnerships. They aim to transform drug development by moving beyond trial and error to a system that learns, adapts, and delivers, ultimately bringing the right therapies to the right patients sooner. Pathos integrates advanced AI technology, biological modeling, and the largest collection of oncology patient data to make clinical trials smarter, faster, and more precise. Their platform, PathOS™, is powered by a shared AI core, Foundry, which drives Scout and Sprint to prioritize promising assets and translate them into faster clinical learning. Foundry learns from every program, ensuring each new asset starts stronger than the last. Pathos was founded in 2022 by Eric Lefkofsky and Ryan Fukushima, with Iker Huerga joining as CEO in 2025.

Products

• PathOS™ platformAn integrated platform that combines advanced AI technology, biological modeling, and a large collection of multimodal oncology patient data to redefine drug development. It is powered by a shared AI core, Foundry, which drives Scout and Sprint engines.
• FoundryThe shared AI core of the PathOS™ platform. Its backbone is the Oncology Foundation Model, built in partnership with Tempus and AstraZeneca, which is the largest model of its kind. Foundry connects to Pathos’ wetlab for a lab-in-the-loop validation system, compounding learning across every program.
• ScoutAn AI-enabled asset selection engine that uses advanced AI to identify therapies most likely to help patients who don’t respond to today’s standard of care. It scans investigational therapies, matches them to patient subgroups, and ranks opportunities for development.
• SprintAn AI-enabled clinical execution engine. Once Scout selects an asset, it moves into a Sprint Pod, a small autonomous team that operates like an independent biotech company inside Pathos. Sprint Pods utilize the PathOS™ platform to design biomarker-driven clinical trials and accelerate Go/No-Go decisions.

Key capabilities

• AI-enabled precision oncology biotech
• Clinical-stage
• Multimodal patient data integration (genomics, transcriptomics, imaging, text, longitudinal clinical outcomes)
• Biological modeling
• Proprietary data model with thousands of AI agents
• Largest collection of oncology patient data (>200 petabytes, 50x TCGA)
• Oncology Foundation Model (in partnership with Tempus and AstraZeneca)
• Lab-in-the-loop validation system
• AI-enabled asset selection (Scout)
• AI-enabled clinical execution (Sprint Pods)
• Continuously learning system for drug development

Use cases

• Redefining drug development
• Accelerating clinical trials
• Increasing success rates in clinical trials
• Identifying patient populations most likely to respond to therapies
• Designing biomarker-driven clinical trials
• De-risking investment in drug development
• Developing precision therapies at scale
• Advancing clinical programs with real patients

AI approach

Pathos is an AI-enabled, clinical-stage biotech focused on precision oncology. They use a proprietary AI platform called PathOS™, which includes Foundry (an Oncology Foundation Model built in partnership with Tempus and AstraZeneca), Scout (an AI-enabled asset selection engine), and Sprint (an AI-enabled clinical execution engine). This platform analyzes a large collection of multimodal oncology patient data (genomics, transcriptomics, imaging, text, longitudinal clinical outcomes) to identify patient populations most likely to respond to therapies, design clinical trials, and accelerate drug development. They emphasize building their own models and integrating AI into every stage of their drug development process.

Tech named: AI, Machine Learning, Multimodal Foundation Model in Oncology, AI agents, Oncology Foundation Model

Industries served

• Biotechnology
• Pharmaceuticals
• Oncology

What it says sets it apart

• Largest collection of oncology patient data (millions of records, >200 petabytes)
• Proprietary data model with thousands of AI agents simulating trial scenarios
• Revolutionary new standard for drug development leading to higher success rates and faster timelines
• Repeatable process that de-risks investment and compounds value
• Building an engine to repeatedly discover and develop precision therapies at scale, not just betting on one drug
• Science-first approach grounded in oncology expertise
• Data-driven with multimodal evidence at scale
• Focus on advancing clinical programs with real patients, not just compressing discovery timelines
• Unique data advantage with multimodal records at unmatched scale
• AI infrastructure that grows smarter with every trial
• Integrated engine pairing model-driven hypotheses with biological validation and biomarker-driven clinical execution
• Oncology Foundation Model built in partnership with Tempus and AstraZeneca

Funding rounds we track

From the AI funding tracker — rounds as reported by the linked publications.