Genialis™ krasID

A platform for custom KRASi predictors,
powered by RNA‑driven machine learning

Based on gene expression data, krasID incorporates over 20 aspects of KRAS-associated biology

  • Identify patients most likely to benefit from your KRAS inhibitor, and understand why
  • Predict the duration of response

  • Understand early disease progression in KRAS-mutated cancers treated with KRAS inhibitors

krasID generates a fundamentally new type of biomarker that reflects underlying biological states, not just genetic variants.

Customizable for different KRAS inhibitors, tissue types, or treatment combinations, krasID can be tuned to match your specific preclinical or clinical context.

Case Study

A krasID predictor accurately stratifies KRASi-eligible patients by clinical response

  • Real World Data
    Patient data from real-world sotorasib-treated cohorts mirror those in the CodeBreaK100/200 clinical trials
  • Longer Time on Treatment
    Patients identified as responders with krasID remained on sotorasib nearly 50% longer than the entire cohort of G12C-selected patients
  • KM plot of predicted benefit to sotorasib in a NSCLC RWE cohort

KM plot of predicted benefit to sotorasib in a NSCLC RWE cohort

Genialis™ krasID outperforms current Standard of Care Biomarkers, which are limited to mutational status

Historical Approach

  • Limited to presence or absence of DNA mutation

  • Perhaps necessary, but utterly insufficient to predict efficacy (mutation selected ORR ~ 30-40%)

  • Cannot inform time on treatment or combination strategies

Genialis™ krasID Advantage

  • Integrates signal from KRAS biology with surrounding tumor milieu using RNA-seq & ML

  • Predicts response with >94% accuracy in preclinical models and >80% accuracy in real world patients

  • Stratifies patients based on time on treatment/survival

  • Reads out actionable changes to other therapeutically relevant biologies

Want to learn more?

Download the Genialis™ krasID White Paper

Explore the full story behind the science and innovation of krasID. Inside, you’ll find:

  • Why RNA Phenomarkers Matter
    How RNA-based, biology-first machine learning overcomes the limitations of DNA-only biomarkers

  • Platform Architecture Unpacked
    How the Genialis™ Supermodel and ResponderID™ work together to tune customizable, interpretable KRASi predictors

  • Validated Real-World Results
    How krasID predictors perform in different preclinical and clinical case studies
  • How to Engage with Us
    Learn how to get started with krasID

More than 50 companies are competing to bring KRAS drugs to market, with the ‘best-in-class’ distinction yet to be claimed.

Gain the edge for clinical trial success and market share by stratifying your patient population with 80% predictability.

We welcome you to explore & evaluate Genialis™ krasID. Schedule a free consultation with our experts to begin!