Noninvasive Stratification of Colon Cancer by Multiplex PET Imaging

Area of Science:

• Oncology
• Molecular Imaging
• Biotechnology

Background:

• Current colon cancer subtyping via gene expression profiling is invasive and lacks dynamic insights.
• Molecular imaging, particularly PET, offers a noninvasive alternative for tumor visualization.
• Key factors influencing PET phenotypes and their correlation with tumor genotype remain underexplored.

Purpose of the Study:

• To investigate the feasibility of noninvasive molecular stratification of colon cancer using PET.
• To identify factors influencing PET imaging phenotypes.
• To explore the relationship between PET phenotypes and tumor genotype/subtypes.

Main Methods:

• Conducted 126 PET scans across five preclinical colon cancer models with diverse genetics.
• Utilized four metabolic PET tracers: [18F]FDG, [18F]FET, [18F]FLT, and [11C]acetate.
• Evaluated tumors at subcutaneous, orthograft, and autochthonous sites, and at primary and metastatic stages.

Main Results:

• PET imaging signatures showed significant heterogeneity, influenced by genotype, tumor environment, and stage.
• [18F]FLT and [18F]FET effectively differentiated tumors with oncogenic Kras and Apc loss, respectively.
• [18F]FDG uptake was impacted by the tumor microenvironment; [18F]FET showed higher uptake in a metastatic model.

Conclusions:

• Established the feasibility of noninvasive molecular stratification using multiplex radiotracer PET.
• Demonstrated that PET imaging signatures are influenced by genotype, tumor environment, and stage.
• Laid the foundation for PET-based subtyping in human cancers for noninvasive molecular diagnosis.