Statistical models to characterize colon tumor stiffness heterogeneity through representative atomic force microscopy...

Area of Science:

• Biophysics
• Computational Biology
• Oncology

Background:

• Cellular physical forces are crucial in biological processes.
• Tumor stiffness is linked to cancer progression, tumorigenicity, and therapy resistance.

Purpose of the Study:

• To integrate atomic force microscopy, statistical modeling, and machine learning in colon cancer.
• To correlate local tissue stiffness maps with clinical and phenotypical parameters.

Main Methods:

• Utilized atomic force microscopy to map colon cancer tissue stiffness.
• Applied statistical modeling to identify associations between stiffness and clinical data.
• Employed machine learning to infer clinical parameters from stiffness data.

Main Results:

• Identified known associations (age, tumor stage) and novel links (RAS mutations, colon localization, DNA repair deficiencies) with stiffness.
• Machine learning models successfully inferred clinical parameters from stiffness data.
• Developed a computational framework for integrated analysis of clinical parameters and stiffness.

Conclusions:

• Tissue stiffness is a significant factor in colon cancer, associated with various clinical and molecular characteristics.
• The integrated computational approach provides a framework for understanding tumor heterogeneity and progression.
• This methodology can advance personalized medicine by linking mechanical properties to clinical outcomes.