CanSeer: a translational methodology for developing personalized cancer models and therapeutics

Area of Science:

• Computational biology
• Systems biology
• Genomics

Background:

• Personalized cancer therapy development is advancing with computational modeling and omics data integration.
• Existing in silico approaches often lack a comprehensive framework for target identification, drug repurposing, response prediction, and combination therapy discovery.
• Lung squamous cell carcinoma (LUSC) presents a need for improved, individualized treatment strategies.

Purpose of the Study:

• To introduce CanSeer, a novel methodology for developing personalized cancer therapeutics.
• To create an all-in-one in silico approach for identifying actionable targets, repurposing FDA-approved drugs, predicting treatment efficacy, and discovering novel drug combinations.
• To demonstrate CanSeer's application using diverse LUSC patient data.

Main Methods:

• CanSeer annotates in silico biomolecular network models using patient-specific genetic alterations and RNA-sequencing data.
• Dynamical network analyses are performed to assess treatment responses.
• The methodology is exemplified using three LUSC patient sample types: paired, unpaired, and cancer-only.

Main Results:

• CanSeer effectively predicts treatment responses for personalized cancer therapeutics.
• The study identified effective repositioned drugs for LUSC treatment.
• Novel LUSC treatment combinations were discovered, including Afuresertib + Palbociclib and Dinaciclib + Trametinib.

Conclusions:

• CanSeer provides a robust methodology for personalized cancer therapy development.
• The approach facilitates the identification of effective drug repositioning and novel combination therapies.
• CanSeer holds significant potential for advancing precision oncology, particularly for LUSC.