Published July 15, 2026
Summary: An artificial intelligence system analyzes cell morphometrics and tumor architecture in biopsies to discover species- and tissue-agnostic cancer biomarkers, allowing researchers to accurately translate drug testing results from animal models directly to human patients. This method, as a New Approach Methodology (NAM), allows the user to select appropriate patient populations for cancer drug trials, increasing efficacy and thus increasing odds of FDA approval.
Applications:
- Pre-clinical drug development
- Clinical trial patient stratification
- Therapeutic drug response prediction
- Precision oncology treatment planning
Advantages/Benefits:
- Improved precision oncology and drug development
- Cross-species and tissue-agnostic applicability
- Superior predictive and prognostic accuracy
- Broad utility across multiple cancer types
Background:
Precision oncology relies on animal models to evaluate new cancer therapies. Consequently, there is a critical need to accurately translate these early biological findings into effective medical treatments for patients. However, current methods suffer from poor cross-species translatability despite unprecedented advances in comprehensive cancer molecular profiling (genomics, transcriptomics, proteomics, and epigenomics). A major limitation is the difficulty of linking molecular variation to emergent cellular phenotypes that dictate disease susceptibility and therapeutic response.
Technology Overview:
Scientists at Berkeley Lab developed an AI framework to discover cross-species and tissue-agnostic cellular morphometric biomarkers (CMBs) as a new avenue to improve translatability. Using this framework, they identified CMBs and CMB tumor subtypes from treatment-naïve needle biopsies of mammary tumors in a genetically diverse Erbb2/Neu mouse model, showing significant correlation with responses to docetaxel treatment. These CMBs were then successfully translated to human patients with breast cancer, ovarian cancer or lung cancer, showing superior predictive/prognostic power to biomarkers and/or machine learning systems specifically optimized in human patients. Co-enrichment analysis revealed significantly conserved biological functions associated with CMBs in both mice and humans, such as cell cycle regulation, underscoring their relevance to tumor biology, treatment response and translatability.
Unlike traditional methods relying on species-specific markers, this AI approach leverages universal morphometric cellular features, enabling direct translation of pre-clinical discoveries into clinical applications.
In human cohorts, the system demonstrated superior predictive and prognostic power compared to conventional biomarkers or machine learning systems optimized solely on human datasets.
Development Stage: TRL 8
Inventors:
Status: Patent pending
Opportunities: Available for licensing and / or collaborative research
